Skip to main content

Full text of "The Classical review"

See other formats 




boli” 
es 


aoe npepee-arpnweees 
a ee ene 








Digitized by the Internet Archive 
in 2008 with funding from 
Microsoft Corporation 


https://archive.org/details/classicalreviewO9clasuoft 








Tae ae 


ae | 








THE 


CLASSICAL 
REVIEW. 








VOLUME IX. 
“af 
London : 


DAVID NUTT, 270 ann 271 STRAND. 
1895, 


. 
Pi - 
~ 
, 
. 
- ' 
. 
‘ “ 
i 
= = 
a = 
7 
. - A 
= ‘ 
4] 
- ~ 
cs - 
a 
eo F > 






RicHarp CLAY AND Sons, LimiTEp, = ; 
; LONDON AND BUNGAY. a ieee 
4 cas on ae 
| cy wire: oY mi 

~ ee | | ae 
me rd OE “ 
" | 3 wa Fo! 


‘. ¥ sae pea a 


TABLE O 


No. 


J. Cook Witson. Testimonia for the 
Text of Aristotle's Nicomachean 
Ethics, for the Metaphysics and for 
the ee iecior Analytics . 

A. H. J. Greenipce. The Bineadire af 
the Provocatio . 

W. E. HEITLAND. 
Lucan ; 

W. T. Lenprvun. 
ton and Pindar 

Re. 'TYRRE1. 
and Pindar . i: A eae a ae 

G. F. Hitt. On Descriptive Names of 
Animals in Greece . : 

T. W. Atten. On Descriptive Names 
of Animals in Greece (the word 


iared on ihe Text ei 
PeeParalloie in VEE 


on Paeallels a Milton 


oT OCS) 0s i ir oe eee 
M. L. Earur. Notes on Euripides’ 
Yee: Er re 
M. L. Earte. Note on Soph. Ant. 117- 
7) dd Se eee 
C. M. Mutyany. "Ga the Enclitic Ve . 
E. H. Mines. Note on ci & dye . . 
A. E. Housman. The aaa ee of 
Propertius ; 
W. W. Merry. Note on gee ie 165 
S. B. Puatner. Note on Lucil. i. 24 
T. Nickuin. Note on Intercalation in 
the Attic Year 
Reviews. 


Schenkl’s Epictetus. J.B. Mayor . 

Schenkl’s Collation of the Bodleian 
MS. W. M. Linpsay 

Lafaye on Catullus. Ropinson ELLIs 

Tyrrell and Purser’s Letters of Cicero, 
vol.iv. G. E. JEans. 2 

Sudeman’s Dialogus of Tacitus. 
FURNEAUX : 


iL 


F 


PAGE 


Woe B. 








CONTENTS. 


le: 
Bennett’s Dialogus of Tacitus. H. 
FURNEAUX : 
Bl: vydes’ Adversaria ras eee djments 
Graecorum 7 ie icorum. LL. Camp- 


BELL 


Herion Ae BB ee ee. oS 


Graves’ Philoctetes. M.L. D’OocE . 

Page’s Aeneid i.—vi. H. Exuer- 
SHAW ; 

Reichel on Honene (ees We 
LEAF 


Butler on the Odyssey. ee PLAvr 

Holm’s History of Greece. Vol. IV. 
and the Translation of Vol. I. 
J. B. Bury : 

Sonnenschein’s Greek Cones J. 

Donovan’. . 

Bérard on the Mythology of ‘Arcadia. 

EK. E, SmrEs . . 

Editions of Hyperides by Kenyon and 
by Blass. J. E. Sanpys 

Belling on Tibullus. J. P. PostGaTe 7 

Hartland on the Legend of Perseus. 

F. B. JEvons : 

Monro on the Modes s Crock 

Music. A Reply . ; 

J. S. PuHILymmoreE and §. G. Owen on 
Statius Stlv. I. vi. 44... 
R. C. Jess. Notice of Sir 
Newton aes < shee a 
G. C. W. Warr. In ‘Merioiaes Caroli 

Brewton. fee sos. 

Archaeology. 

JANE E. HArRIson on some points in 
Furtwingler’s theories regarding 
the Parthenon, and on the transla- 
hlonrby b Sellers 5 Oe fee 

| Monthly Record. . . ppeoreny eat 
Summaries of Periodicals . 


ACelas 


PAGE 


85 
93 


iv 


No. 


PAGE 
J. B. Mayor. Critical Notes on the 
Stromateis of Clement of Alexandria 97 
Ropryson Extis. Some Emendations of 


the Greek Tragici el OD 


J. B. Bury. Notes on some Be ey in 
Arist. "AQ. IIoA. . . . 106 
K. P. Harrincron. Notes on Tibull. if 
ny ae elu ie: 
W. even NER. en on ( stston. Were, 
45. fepwtre LOG 
1 3 ei H. Nate on Hor, “Oud. ' iv, 2 49 ..110 
E. J. Cutnnock. Note on 6epid.ov ; LLO 


Reviews. 
Summers on the Argonautica of Val- 
erius Flaccus. R.C. SEATON . 
Pauly-Wissowa’s canes Je 


mG 


HK. SANDYS. AS oe Wy bs: 
Streitberg on W ord: ogee P. GitEs 115 
Merry’s Edition of the Wasps. W.4J. 

IM. STARKIE. . . sie Lil 
Graves’ Edition of the Wasps. ‘E. S. 

THOMPSON peck a 
Strachan- Davidean! s ange oe. Catve, 

L. C. Purser 123 


No. 


J. Donovan. On Greek Jussives . 145 


W: E. Heittanp. Notes on the Text 
of Lucan . . 149 
R. G. Bury. Notes on two passages 1n ) 


Lucretius and in Varro. 
Reviews. 
Hilberg on the Ovidian Pentameter. 
Ropinson ELLIS... Belay 
Editions of Claudian by Birt and by 


. 156 


Koch. J.P. PostGate . ; . 162 
Hauvette on Herodotus. Hvenes 

ABBOTT . . 169 
Hadley’s Edition of the Heouba, #. 

B. ENGLAND: . ara 
Kock’s Edition of the igs: E ‘ 

HirtzeL . . ae I ere 


Sibler’s Edition of the ‘Protayoras, J. 
a eOM GHGs: ec oo) ck teats fe ERNE 


THE CLASSICAL REVIEW. 


2. 


PAGE 
Freese’s Translation of Isocrates. 
W. WAYTE . ae os, 
Hoess on the Style ee ideseutae Ete 
CLARKE 126 


Hultsch on the Tenses “of Polybius. 
E. 8. SHUCKBURGH : keyed 
Hardy’s Christianity and the Roman 
Government. F.C. ConYBEARE. . 129 
Holder’s Scholia Antiqua in Horatium, 


and Friedrich on Horace. T. E. 
PaGe (so sos gjim «se 
J. P. Posreare. The Manuscripts of 
Propertius ts G Sees 
Archaeology. 
Prof. Christ on the Greek Stage. E. 
Capps .. ee 
Reinach’s Cat alogue of ay a Bronces 
in the Musée ne St. Germain. E. 
SELLERS . 136 


Walton on ‘the Cult of Astlepios, 


J. EH. HarRIson smallesye) 





Monthly Record . : : . 138 
Summaries of Purigdicals : . 140 
Bibliography, “./ ec. pene . 142 


3. 


De Mirmont’s Apel et acs 


R. C. SEATON : ae) 
Crusius on the Delphic Hymns G. 
Torr . oie, 
J. P. Postcare, On the MSS, of Pro- 
pertius (A Reply) . . 178 
Archaeology. 
Schilling on the Legions I. Minervia 


and XXX. Ulpia. L. C. Purssr . 186 
Baudrillart on the Goddess of Vic- 


tory. Jane E. Harrison . Fae oi 
Monthly Record. . . é sgaepaled 
Summaries of Pannen ee Peery ric 
ALFRED GUDEMAN. Notice of ine ASS 
Merriam ee A 189 
Bibliography eotagl k 


| 


TABLE OF CONTENTS. v 


No. 4. 


PAGE 


W. E. HeIrLanp. 
Lucan .. 
M. L. Ear.e. 
26-45 52° S.. ; 
J. B. Mayor. eridieal: Rois on the 
Stromateis of Clement of Alexandria, 


Notes on the Text of 

ee eet ees As! LO 
On 5 as Trachiniue 
200 


Book v. . 202 
Reviews. 

Brieger’s Edition of Lucretius. J. 
Masson Saree! 1207 

The Schneidewin- Naudk oe eee 
and #lectra. Epwarp Carrs . . 211 

Butcher’s Poetics of Aristotle. HeER- 
BERT RICHARDS . 2138 


Hilprecht’s Assyrica. G. A. Barton 215 

Pais’ Storia della Sicilia e della Magna 
Grecia. E.S. SHucksurcH .. . 217 

Horton-Smith’s Conditional Sentences. 
K. A. SONNENSCHEIN 

Zoeller on Roman Oaaetutzonal caw 
AL ed. Greenipge . ss ). . 2. 223 


. 220 


PAGE 
Pater’s Greek Studies. A. W. VER- 
RALL . . 220 
Cagnat’s Antiquités Romeines. ay S. R. 229 
Ramsay’s s Roman Antiquities. J 8S. R. 250 
Damsté’s Zectiones Curtianae W. 
©. SUMMERS .. =. . 230 
Blake’s Edition of the "Hallenita: 
R. C. SEATON : rece 
Schenkl’s Edition of ‘Epictanes (A 


Reply.) H. Scnenkn ... . . 231 
Rejoinder to the Above. J. B. 
Mayor . 234 
Archaeology. 
Ruggiero’s Dizionario Epigrafico. F. 
HAVERFIELD. . . . 236 
Stuart Jones’ Select Passages on 


G. F. Hit. 
Monthly Record... . 


Greek Sculpture. 236 

diag aaT 

Summaries of Periodicals . . .. . . 238 
240 


Bibliography . .. + + ee ees 


| 


| 


No, 5: 


A. C. Crank. The Fictitious MSS. of 
Bosius’ . . 241.) 
F. B. JEvons. Saale ‘Sigial Lee aad 
Folklore ee tet a! te . - 24 
W. Lock. On the use of ie in 


E. A. Apsorr. Notes on some Pas- 
sages in Lightfoot’s Biblical Essays . : 
F. C. Conyspeare. On the ore of 
Actsi. 18 in Papias . . . 208 
A. H. J. GREENIDGE. ae aa Title 
‘Quaestor Pro Praetore’ 
S. B. Puatner. Notes on Pian and 
Momentum . 1. + +s eevee 
Reviews. 
Stadtmiiller’s Edition of the Palatine 
Anthology. J. W.M... . 261 
Marchant’s Edition of Thucydides 
vii. CO. Forster Smite ... . 262 


259 | 


Aristotle’s Poetics . ......-. 201 | 


_ Archaeology. 





Edition of the Catiline 
Orations. S.G.OwEnN .... . 263 

Torr’s Ancient Ships. W. RipcEeway . 265 

Churech’s Historical and Political Odes 


Wilkins’ 


of Horace. LL, C. Purser. Sy i 
Tozer’s Selections from Strabo. J. ne 
SU STM RE Tass oss cists Se . 268 


A. Furtwancier. On the Lemnia of 
Pheidias and the Parthenon Sculp- 


tures. « es 
CrciL SMITH. On the “Myth of Txion 277 
KE. E. Sixes. On Nike and Athena 
WNiké «4.7 . 280 
Monthly Bacord.. widotues . 283 
Summaries of Pariodieala : . 284 
Bibliography . 287 


vi 


INO: 


PAGE 
J. Donovan. German Opinion on Greek 
Jussives . 289 
C. D. CHAMBERS. rhe Glaisahennan is 
Conditional Sentences . . 295 
F. C. ConysBeare. On the Inseviption 
of Abercius . 3 . 295 
Jue MAvOR. Critical Wales on fe 


Stromateis of Clement of Alexandria, 


Book vi. : . 297 
J. Gow. Hor: Shane B : SOO 
H. Ricuarps. Catulliana . a04 


E. S. Tuompson. Notes on the Wasps 
of Aristophanes . 

A. G. Hopxins. Ona passage in the 
Trinummus of Plautus . 


. 306 


. 307 


EK. C. Marcuant. - Note on n Thueyd, Vi. 
1 Sate ore . 009 
F, HAvERFIELD. Note on Tac. Dose 
ae kOe, 
E. Riess. von ae ord iagnaiior as 
asa point of latitude .. . Sau 
C. Courr. Note on the Homer Digee 
resis . ee 
No 
J. B. Mayor. Critical Notes on the | 


Stromateis of Clement of Alexandria, 
‘ Book vi. : Sey 
J. Donovan. German “Opinion on 





Greek Jussives (continued) . 042 
8S. G. Owen. Notes on Juvenal . d46 
K. H. Donkin. ék or dwd denoting 

PPR yg ee ing) SV ee, Nee ener 
Reviews. 

Postgate’s Edition of Propertius. 


A, E. Housman z . 050 
Onions’ Edition of Nonius Manosllae 
W. M. Linpsay ; . 056 
Sudhaus’ Edition of Pisledenties ae 
EK. SANDYS .. 
Forbes’ Edition of Thucydides. 
i. E. C. Marcnant : 
Rogers’ Emendations in Greek Tragic 
Poets, L. CAMPBELL . . 362 


Ws) fy 
"tae 
. 360 


THE CLASSICAL REVIEW. 


6. 


PAGE 
F. T. Cotzy. Note on Matt. xi.19 . 312 
Reviews. 
Waddell’s Edition of the Parmenides 


of Plato. R. D. Hicw® .. 3 eee 
Hilgard’s Grammatici yraeci, Part 
iv. (Choeroboscus). E. G. Srmmer 317 
Corpus Poetarum Latinorum, Fascte. 
ii. T. W. JacKson 
Furneaux’ Edition of the Gerinanea 
of Tacitus. W. PETERSON ea26 
Davis’ Edition of the Germania, and 
Stephenson’s of the Agricola and 
Germania. W. PETERSON. . . 326 
Reid’s Edition of the Pro Malone of 
Cicero. G. G. Ramsay . 800 
Maurenbrecher’s Carminum Suiariin 


Reliquiae. KR. 8. Conway . 302 
Archaeology. 

Catalogue of Greek Coins of Troas, 
heohe: and Lesbos. W. RIpGE- 
WAY .. eg. REE, oo vie . 330 

Monthly Record . . 335 
Bibliography . 336 


Fle 


Holden’s Edition of Plutarch’s Peri- 
cles. ¥. A. Hintzun. : 360 
Papillon and H aigh’s Text Gg Vireil 


S. G. OwEN . . . 366 
Melber’s Edition ae Wios Caseig: G. 
MecN. Russrorty . ween 


Madan’s Catalogue of Westen MSS. 
in the Bodleian: Roginson Evxiis 367 
Baker’s Latin and Greek Translations. 
E. D. Stone . 069 
Archaeology. 
On the Thymele in Greek Theatres. 


A. Bernarp Cook . 2 geen 
Professor Ridgeway’s Review of 
Torr’s Ancient Ships.. Cxrcit Torr 

and W. Ripceway . . 378 

Monthly Record . . 319 

| Summaries . 380 

| Bibliography . 382 


TABLE OF CONTENTS. 


No. 


PAGE 


J. B. Mayor. Critical Notes on the 
Stromateis of Clement of oe 
Book vii. 


G. C. W. Wine. * The PPasiodia Hccntess 


L. R. Hieeins. On the meaning of 
BovAopacin Homer. . . Poe: | 
M. L. Earue. Notes on Soph. Trach. 


56 and Eur. Med. 13 . 395 


J. Woop Brown. The Geinceiions in 
the Florence MS. of Nonius . 396 | 
Reviews. 
Lindsay’s Latin Language. R. &. 
Conway . 403 
Verrall’s Euripides the Rationalist. 
J. R. Moziey . . 407 


| 


vii 
8. 
PAGE 
Sehrwald’s che bios E. E. 
SIKES es 
Gehring’s Jndex Hoots: Wwe 
ALLEN ; DAD 


The Oracles Aseribed ‘to Matthew by 
Papias, a Contribution to the Critic- 


ism of the New Testament. A. C. 
HEADLAM . . 419 
Archaeology. 
The ‘System’ in Greek Music. C. F. 
Appy WILLIAMS . ae “etek 
Note on the Parthenon. JANE E. 
Harrison . 427 
Monthly Record .. . 428 
Summaries . 429 
Bibliography . 431 


No: 9: 


J. B. Mayor. Critical Notes on the 
Stromateis of Clement of Alexandria, 
Book vii. (concluded) . 433 

M. L. Earrze. Miscellanea Critica . 

H. F. Pernam. The Emperor Claudius 
and the Chiefs of the Aedui . waa! 

R. Exuis. Geographical Notes on Pro- 
pertius . : 

J. Donovan. German. Opinion c on Gr eek 
Jussives (concluded) ess 

J. Woop Brown. The Corrects in 
the Florence MS. of Nonius . 

Reviews. 

Susemihl and Hicks’ Edition of the 
Politics. . J. A. STEWART . ; 
Smyth’s Jonic Dialect. P. GILEs . 


Buck’s Oscan-Umbrian noe 
EORRTED “se: . 460 
Cooper’s Word-For mation in , the Roman 


Sermo Plebeius. H. W. Haytey . 462 
Cauer on the Groundwork of Homeric 
Criticism. W. Lear . . 463 


. 439 | 


2 444 





Wattenbach’s Introduction to Greek 


Palaeography. F.G. Kenyon . . 465 
Wachsmuth’s Introduction to the 
Study of Ancient tie F. 
HAVERFIELD . : . 466 
| Archaeology. 


American School of Classical Studies 
in Rome. G. E. Marinpin : 
Weil and Reinach’s Edition of ae 


Second Delphian Hymn. D. B, 
MonRO::.. . . 467 

Navarre on the Gisak Thestte: a 
RY RAIGH. 6.0 . 470 

Foucart on the Eleusinian “My steries. 
AB SES i rk ds, . 473 

W. WarRDE FowLer, Was ike Flam- 
inica Dialis Priestess of Juno?. . 474 

Cecir Torr. Ancient Ships. (A 
Reply) . . Nee Sie oe dO 
Summaries ATT 
Bibliography 478 





The Classical Review 


FEBRUARY 1895. 


TESTIMONIA FOR THE TEXT OF 


FOR THE METAPHYSICS AND FOR THE / 


THE pabypatixi) (or peyddyn) ovvtagis of 
Claudius Ptolemaeus and the commentary 
upon it by Theon of Alexandria contain 
references to the matter of passages in the 
second and sixth books of the Nicomachean 
Ethics. 

N. Eth. II. i. 1. 1103* 14. derris 8 rips 
apeTas ovans THs pev StavontiKns THs de HOLuKHs, 
H pev StavontiKi TO wAEtov ex SidacKadias €exeEL 
Kal Ty yeveow kal THY avéyow, di7Ep ep- 
Teiplas Seirat Kal xpovov, 7 Oe 7OiKH €& EOous 
repiylyverat, bev Kal Totvopa eoynKEv puKpOV 
mapekkNivov dao Tov eOovs. 

Cl. Ptol. pey. ovv. wpooimiov, init. rave 
KadOs of yvyociws giAocodotvtes, © Zuvpe, 
Soxotat por Kexwpikévar TO OewpytiKov Tips 
prrocodias. kal yap ei ovpBeBnxe Kal TO 
TPaxTiKa mporepov avTov TovToU Gewpytixov 
Tuyxave ovdey WTTov dv Tis evpor peyddAnv 
otoay év abrois Ouaopav: ob povov dua TO TOV 
pe Ouav dperov eévias imdpéar divacba 
ToAXois Kal xwpis pabjoews THs b€ TOV dAwV 
Gewpias advvarov <ivat TvxEly avEv diwWacKaAtas, 
GAXG. kal TG Tiv TAcioTHV Opediav exer eV EK 
Tis €v avTOLS TOIs TpdypacL GvVEXOUS EvEpyeElas, 
evade de ex THS ev Tos Hewpdpaci TpoKoTTs 
mreprylyveo Gan. 

The corresponding part of Theon’s com- 
mentary (p. 1, Basle edn. 1538 ; p. 3 bottom, 
of Halma’s edition) is as follows :— 

dyot be 6 IroAepaios ovpBeBnxevar TO 
TPaKTiKg TO mporepov avTov TO Gewpyrixdy 
TvyXaVeELV, bud 70 iows beiv ™porepov TOV Tpd- 
favta te Kal Ore aiperov TO mpaxOnoopevov 
KateiAndevat, kat OTe dur TOVOE GV yevoito Kal 
Tovde TOV TpdTOV, arep eotiv GANnOevTiKAS Kal 

NO, LXXV. VOL, IX. 


ARISTOTLE’S VICOMACHEAN 


ETHICS, 
POSTERIOR ANALYTICS. 


Gewpytixi)s efews" GAN opus pyai peyadny 
eivar ev avTots TV drapopav- TOV pev ‘yap 
HOuKOV dpetov evias Kat avev didacKadias 
meprylyverOar, €€ EOovs yap dracat avra. wAqV 
dpovycews doxotor. cvvictacba, Obey Kal 
nOiKas avTas agvotow dvonaleorbat, otov éOixas 
twas ovcas. ici & attrac cwhpootvyn avdpia 
eAevGepioryns Sixaoovvyn mpactys’ Kal amdOs 
Kahol Kat ayaboi +o os civar deyopucba. 
doxotor O€ TO’TwY TIVés Kal volLK@s Tapayi- 
veoGa.. kal yop Kal ddoya loa Ta pev avdpeta 
Tu b€ THppova €yerat elvan. 

The passage from Theon may be counted 
as independent testimony, for though he 
only refers to Ptolemy he is nearer the 
Aristotelian text than Ptolemy. It will be 
seen that Theon’s additions are not put 
in oratio obliqua like some of the sentences 
he takes from Ptolemy, though it is un- 
necessary to lay stress upon this. ‘The 
latter part of the passage from Theon seems 
to refer to the doctrine of Vic. Eth. V1. 
xiii. 1, 1144” 4, wéow yap doxet exaora Tov 
nOav imdpxew pice Tus (Kal yap Adtkavor Kal 
ocwdpovixol kai avdpetor kat TAAAG Exopmev Edis 
ek yeverns)’ GAN’ dpws yyovpel’ Erepov te TO 
Kupios dyabov Kal Ta Towatta GAXov TpoTov 
dmdpxew. Kal yep Traut Kat Onptous at dvovxal 
bmdpxovcw éfets. 

There are several points in these various 
passages and in their relation to one another 
which seem worthy of discussion. 


I, 


It is not absolutely certain that Ptolemy 
had the Aristotelian text before him. If 
B 


2 THE CLASSICAL REVIEW. 


évias Tov 7OuKav aperov implies, as Theon sup- 
poses, that Ppdvyois is counted among the 
iOuxai dperat, there is an important depar- 
ture from Aristotle’s doctrine, though it is 
such a one as the difficulty of his own 
representation was very likely to occasion : 
and it is at least not impossible that this 
change, as well as some minor ones of 
expression, is not due to Ptolemy himself 
but to a later Peripatetic version of the 
passage. A little lower down Ptolemy quotes 
Aristotle by name (Halma p. 2) kal yap at 
Kal TO Gewpntixov 6 ’ApiototéAns wavy epperd@s 
eis Tpia TX mpOTa yevyn Siaiper TO TE PvoLKoV 
Kal TO pabypatiKov Kal Td OeoAoyiKor, K.T.X. 
This distinction is found in Metaph. x, 7 
(cf. 1064” 2), and in the parallel version in 
ce, 1 (cf. 1026* 19); but Ptolemy’s account 
of it though mainly Aristotelian (cf. eg. 
Metaph. X 1072* 26) differs a good deal in 
form from that in the Metaphysics and seems 
to show the influence of later ideas. 


(HE 


If the passage above quoted from the 
Nicomachean Ethics (1103* 14) be compared 
with the corresponding statements in the 
second book of the Hudenian Ethics (1220* 
39) and in the Magna Moralia (1185° 38), it 
will be clear that of the three it is the 
Nicomachean version which, whether 
directly or through a medium, is the proto- 
type of Ptolemy’s quotation: and the same 
is true of the quotation in Theon’s com- 
mentary. Again the reference peculiar 
to Theon is obviously nearer to the 
Nicomachean Ethics (1144” 4) than to the 
version of the same thing in the Magna 
Moralia (1197° 38). Now when an ancient 
writer refers, as Theon does here, to an 
undisputed Nicomachean book, and at the 
same time to the matter of a passage in a 
disputed book, it becomes important to 
observe whether the latter reference seems 
to be to our text; for, if it were, as it is 
rather more probable than not that both 
the quotations would be taken from the 
same version of the Hthics, the hypothesis 
that the disputed book is Nicomachean 
would be somewhat strengthened. Here 
however the resemblance is not close enough 
to make it certain that the extant version 
of the disputed book is the one quoted, nor 
remote enough to make it probable that the 
reference is to the other version (i.e, Lud. 
or Nic.), if indeed such other version ever 
existed.) 


1 ; ; 
_ | Here two questions may be asked. Is any 
instance known where a disputed book is quoted in 


EYE; 

The passage from Ptolemy may rank as 
one of the earliest of those quotations from 
the Hthics of which the date can be approxi- 
imately fixed. It may be earlier than the 
commentary of Aspasius, and is at least 
about contemporary, for Ptolemy -and 
Aspasius are both said to have flourished 
about 125 a.p. And when it is remembered 
that Proclus belongs to the fifth century 
A.D. and that Simplicius and Philoponus 
are as late as the sixth century, the refer- 
ence in Theon of Alexandria gains in 
importance, for he is said to have flourished 
in the latter half of the fourth cen- 


tury A.D. 
Li fs 


One may venture to think that in the 
passage above quoted from Vic. Hth. IL. i. 1 
there is a difficulty about the position of 
the words did7ep éeureipias Setrar Kat ypovov 
which makes the testimonia of interest. 
It is true that in Mic. Hth. VI. viii. 5, 
1142* 11—16 ¢povycis is said to need ex- 
perience and time (7zA7Oos yap xpdvov rrovet 
Tv éumepiav), and in the next passage,” 
1142* 16—20, the same is said of codia, 
dpovncts and godia being duavontixal dperat. 
But according to the same two passages 
mathematics is contrasted with codia and 
dppovno.s as not needing experience, and 
mathematics is the conspicuous instance 
in Aristotle of a science which proceeds by 
didacKadia, in the technical sense. Hence 
the fact that a science proceeds by divdacKxadia 
could not be a reason for its needing time 
and experience, as affirmed in the received 


antiquity under the title Hudemian Ethics? Does 
any quotation of the matter of the disputed books 
point to a version different to the extant version ? 
As far as the list goes which is prefixed to Susemihl’s 
edition (cf. also Fritzsche’s) the answer to both ques- 
tions seems to be in the negative. And it is worthy 
of note that the undisputed EKudemian books are 
sometimes quoted under their title ‘Eudemian,’ 
while the disputed books are quoted as Aristotle, 
Aristotle’s Ethics, or the Ethics, oras the Ethics with 
the number of the book given as it is in the 
Nicomachean version, and not as in the Eudemian, 
or lastly with the definite title of Nicomachean Ethics. 

* The two passages, though contiguous, are dis- 
tinguished in the above because there are peculiarities 
in the form of the second which suggest that it may 
be a later addition to the first. If this were so, it 
might account for the apparently inaccurate use of 
motevovar for érlorarra (T& MEV Ov MiaTEVOUEL Of VEoL 
GAAG A€youot): for notwithstanding the wide sense 
of miorevew, this is just a case where it ought to 
be distinguished from éricrac@u. The difficulty is 
removed by Imelmann’s ingenious emendation (Ta wey 
mirTevouoty of véot %AAos A€youoww) : but the form of 
the received text is somewhat confirmed by Nie. Eth. 
VII. iii. 8, 1147* 18 —22. 





THE CLASSICAL REVIEW. 3 


text of Nic. #Hth. Ll. i. 1. On the other 
hand, it is characteristic of the moral 
virtues that they require time and ex- 
perience because they come through habitua- 
tion, and it is the very object of the second 
book of the Vic. Ethics to make this clear. 
We should expect therefore rather 7 pév 
diavoyntixy) TO mAciov éx didacKadias exer Kat 
Ti yeveow Kal THY avéqoy, 1 Oe 76K) e& EOovs 
meprytyverat (d0ev Kal Tovvopa exynKey piKpov 
mapexk\ivovy ard Tod eOovs) dudrep eprretpias 
detrar kal xpovov. 

The Paraphrast evidently found the text 
as we have it, but his date is probably so 
modern that this is not of much consequence. 
Aspasius does not quote or paraphrase the 
suspected words at all ; yet he has a remark 
which suggests that he may have possibly 
found them where they are and felt a 
diticulty about them. He says «i 8 dpa 
pederns twos det ev TO pavOdvev, Toito 8 «i 
BotXerat tis KaXciv €fos, kadeitw. GAA TO ye 
Kuptws €Jos é€oti TO eOiLerGar Kadots érern- 
devpac. Kal piv » Ppovnos dua woAARS ep- 
Tetplas tapaylyverat Kal dudacKkadtas. It looks 
as if he felt that éeuepia and fos were 
more distinctive of 76x) dpery as such, 
though conceding on the one hand that in 
the limited sense of peAérn they belong to 
the diavonrixal dperai in general, and on the 
other hand that dpdvyots in particular, which 
is one of these virtues, needs, beside d.dac- 
kadia, also éyzepia in the proper sense. 
However this may be one can hardly found 
an argument for or against the position of 
the words di67ep x.7.X. on Aspasius. 

In the passage from Ptolemy the expres- 
sion nearest to du7ep x.t.X. IS ek THs ev 
avtois Tois Tpaypac. ovvEexos evepyeias, and 
this is connected with the ethic, not the 
dianoetic virtues. But it is quite possible 
that the words are only an expansion of 
e€ €Gous. 

Theon has nothing which might corre- 
spond to the difficult words or indicate 
_ where he found them, if he found them at 
all. 


LF 


Even if the testimonia were favourable 
to a transposition, the variations either in 
commentary or quotation have to be used 
with caution, especially when the difficulty 
in the original arises from the connexion of 
the argument. The ancient writer may 
have felt the same kind of difficulty about 
the original passage as the modern student, 
and consciously or unconsciously may have 
departed from the original. A warning 


example is given by the manner in which 
Proclus reproduces a certain passage of the 
Organon. The original is: dxpiBeorépa 8 
ETLITHLN ETLITHUNS Kal TpoTépa, 7 TE TOD STL 
Kat dvdte 7 adTH GAAG pa xwpis Tod Ste TAS TOD 
dior, Kat 7) py Ka broKewevov tas Kab’ tiro- 
KELJLEVOU...- kat 9 €€ €AatTOvwy Tis éx poc- 
Oéoews x.t.A. Post. An. 87* 30. There is 
here a known difficulty. From what is said 
elsewhere in the same treatise, 78" 32 sqq., 
and from the sequel of the passage itself, 
the reader would expect to find that the 
science of the du0rr alone would be ranked 
higher in exactness than the science which 
combines the édc67c with the 67, and this 
again higher than the science of the 6rt. 
Yet the statement of the passage is clear, 
and there can be no reasonable suspicion 
that the words are corrupt. No plausible 
emendation—in fact nothing less than an 
entire rewriting of the passage would give 
the sense expected. Proclus (Comm. in pr. 
Lucl. Element. lib., Friedlein p. 59 1. 11) 
writes: dxpiBeorépa ydp éotw émurtipyn GAA 
aAdXys, os pnow ApiotoréAns, y Te && dm\ove- 
Tépwv trodécewv Opunpéevn THs Tokiwrépats 
Gpxats xpwpevns! Kai 9 Td dT. A€youvca Tis 
TO OTL yryvwoKovons K.T.A. 

He seems merely to have avoided the 
hard words, making the passage easy by 
substituting for them the kind of formula 
which suits the doctrine of 78° 32. In fact, 
as may be seen by referring to the con- 
tinuation of the passage partly quoted above, 
Proclus combines Post. An. 87* 30 with 78” 
32 as if there was no difference between the 
two places.” 


1 The text has % re woutAwrépais apxais xpwuévn 
THs ek awAovoTEépwy brobésewv wpunuevns. The cor- 
rection is due to Barocius (c7#. Friedlein) and seems 
obviously right, for the phrase corresponds to  é 
éAattévwv Tis éx mpoocbécews in the original. 

2 In the same treatise of Proclus are three refer- 
ences to the Ethics: in p. 32, 1. 4 (Fried.) to Nic. 
Eth. 1095° 1, in p. 33, 1. 25 and p. 192, 1. 10 
to Nic. Eth. 1094v 26. These do not occur in 
the above mentioned list of testimonia, though they 
come from a book of which there is a modern edition 
with a copious index ; it may be suspected therefore 
that the list is capable of a good deal of extension. 
There are many references to Plato in the same book, 
and especially to the 7'imaeus, which are interesting 
sometimes as confirming curious expressions in Plato’s 
text, or as contributing to interpretation (cf. e.g. 
Fr. 20, 10, Tim. 58 C; Fr. 562, 20, Tim. 42 A; 
Fried. 108, 10, Tim. 42 E; Fr. 291, 1, Zim. 87B; 
Fr. 382, 3, Tim. 53 C). It contains references to 
passages later than the part of the 7imaeus at which 
Proclus’ commentary ceases, The Commentary is of 
course often quoted, but the editors of the Zimaeus 
seem to have made little or no use of this other 
source. Stallbaum quotes it perhaps not more than 
once, and then through the medium of Boeckh, and 
on a historical point, not for the text. 

B2 


way 


The passage from Theon’s commentary 
seems to require emendation. The clause 
kat dmh@s Kadot Kat dyafot 1d os ctvar 
AeyopeOa might be construed as an inde- 
pendent sentence with the emphasis on 
€Gos. Yet this would be very harsh: the 
sentence doxodtou d€ TovTwy twés would be 
separated awkwardly from the list of moral 
virtues to which it directly refers. Again, 
that list would be naturally terminated by 
a general expression to cover any virtues 
not enumerated, and kai dwA@s is a phrase 
by which such an expression is properly 
introduced. Hence it may be inferred that 
something has dropped out between amas 
and xadoi, perhaps ais absorbed by the 
termination of a7Ads, or xa’ ds lost through 
similarity of initial syllable to that of xadoé. 
The sense also seems clearly to require 700s 
for €Gos in the same clause—a conjecture 
confirmed by a passage a little further on 
(Halma p. 4) pds ye pay tiv Kara Tas 
mpages Kal TO 7) 00s KaAokayabiav. The part 
emended would then read thus: dev kat 
nOukas aitas agvodow dvopaterOar otov ebiKas 


THE PROCEDURE OF 


‘THERE is probably no subject connected 
with Roman criminal procedure about which 
such vagueness prevails even at the present 
day as that of the exact nature of the 
‘provocatio ad populum.’ To the student 
of courts of appeal the question of main 
interest must always be whether the ‘ pro- 
vocatio’ was a true appeal; that is, 
whether the people could, by this procedure 
amend, as well as confirm or reject, a 
sentence. Under great varieties of state- 
ment we find a general agreement amongst 
modern authorities that the people possessed 
this power. Mommsen (Staatsrecht ii. p. 
978 note 3) says that the ‘ provocatio ’ 
goes from the magistrate to the ‘ comitia,’ 
and is not merely ‘cassatory’ but also 
‘veformatory’; Merkel (Uber die Geschichte 
der Klassischen Appellation) thinks that, at 
least in the case of ‘multae,’ it may have 
been reformatory ; in Smith’s Dictionary of 
Antiquities (s.v. ‘appellatio’) we read that 
‘the “provocatio’”? was an appeal in the 
strict sense of the term, 7.e. it consisted of 
a rehearing of a case previously tried and 
a new judgment upon it’; and this belief 


THE CLASSICAL REVIEW. 


Twas ovoas. eiot 8 attra: cwdpooivy davdpia 
éhevbepiorns Otkaocvvn mpadtys, Kat amos 
<ais> xadot kat adyafol To 700s elvar deyo- 
pba Soxotor dé TovTwy Ties K.T.A. 

It may be noticed that the conception of 
Kaos kat dyafos here and of kadoxayabia 
seems to be of the general kind found in 
the Nicomachean Ethics and the Politics: 
for xadoxaya6ia in that special sense which 
is peculiar to the Hudemian £¢thics includes 
dianoetic as well as ethic virtue. On the 
Eudemian view therefore it ought not to 
appear at all in a list of ethic virtues, such 


-as Theon here gives, and ina general list of 


the virtues would pretty certainly be repre- 
sented by the substantive (kadoxayafia) and 
not by the adjectives as above. In the 
second passage from Theon xadoxayaOia 
appears, not as one among other virtues, 
but as designating moral excellence in 
general (7) kara Tas mpagers Kal TO G05 KadoKa.- 
yabia) just as it does in Politics 1259” 34 
and Nic. Hthics 1179° 10. 

In the second line of the first passage 
from Theon the editions have xaradnévat ; 
Halma gives it without variant. 

J. Cook WILson. 


THE *‘ PROVOCATIO.’ 


even underlies the apparently contradictory 
statement of Marcel Fournier (Zssaz sur 
Vhistoire du droit d’appel, p. 40) that ‘the 
* provocatio ”’ did not tend to the reforma- 
tion of a sentence like the appeal, it 
changed the competent tribunal, which 
permitted a new judgment to be rendered 
that had no connexion with the first’ ; for, 
where a tribunal is changed after a sentence, 
there we have the true appeal. But no 
adequate explanation is vouchsafed by any 
of these authorities as to how this reforma- 


tory character wosattained. Anexplanation ~_ 


could only be furnished by an accurate 
knowledge of the procedure of the 
‘provocatio’; but here we are met by 
the initial difficulty that, as Geib says 
(Criminalprozesse, p. 168), nothing is known 
about such procedure. This is literally 
correct ; with the exception of the brief 
account, meant to be typical, of the trial of 
Horatius, no description of a ‘ provocatio’ 
has been preserved in Roman history, unless 
the trial of Rabirius for ‘ perduellio’ can be 
considered a true case. The reason for 
considering it to be one is that the procedure 





THE CLASSICAL REVIEW. 5 


here was consciously antiquated. The 
duumviri were appointed by the praetor (as 
originally by the king) and not by the 
people, and hence they could be appealed 
from ; for the popular creation of duumviri 
at this period would no doubt have followed 
the analogy of the appointment of special 
quaesitores and have, therefore, excluded 
the appeal. At an earlier period the 
duumviri had probably been a magistracy 
(Mommsen, Staatsrecht ii. p. 616) and as 
such technically subject to appeal. Conse- 
quently, the only mode of answering the 
question is to assume, as is usually done, 
that procedure in ‘provocatio’ was, or 
became, identical with procedure in a 
‘ judicium populi.’ 

The connexion between the two is 
undoubted ; but various explanations have 
been given as to what this connexion was, 
explanations differing according to the view 
taken of the original theory of criminal 
appeal. According to Ihering (Geist des 
Rémischen Rechts, I. p. 257 ff.) the ‘ judicia 
populi’ are historically independent of the 
‘provocatio. The latter is a denial of 
competence of a magistrate, not an appeal 
for pardon so much as ‘a protest of nullity 
on the ground of incompetence,’ and _pre- 
sumes, therefore, from the first the 
preexistence of popular courts with fully 
admitted spheres of jurisdiction. The 
logical consequence of this would be identity 
in procedure, the only difference being that 
in some cases the word ‘provoco’ must be 
used before the trial, in others it would not 
be used. In both courts the king presides : 
but the one is a court of first instance, the 
other a court of appeal. 

According to the more general view, the 
‘judicia populi’ were developed from the 
‘ provocatio,’ which was originally an appeal 
for pardon. If, in pursuance of this view, 
we hold that the latter became merged in 
the former, it is a natural, though by no 
means an inevitable, conclusion that the 
_ procedure by appeal remained a constant 
element in popular jurisdiction. This 
theory, first fully formulated by Mommsen, 
may be briefly stated in the form that the 
‘judicia populi, as they grew out of the 
‘provocatio,’ always continued to be courts of 
appeal. 

But this clearly does not exhaust all the 
alternatives. It is possible that the 
‘ judicia populi’ may have grown out of the 
‘provocatio’ and yet that the two may have 
been kept distinct. There was an obvious 
motive tor a continued distinction between 
the two kinds of courts ; for a trial before 


admittedly no 


the people is the result of a magistrate’s 
recognizing his competence, the ‘ provoeatio ’ 
is the result of his not recognizing the 
limits of his judicial power. There may, 
therefore, have been some similarity in 
procedure, due to their original connexion 
or to later assimilation, but both in theory 
and in practice the popular courts of first 
instance may have been kept distant from 
the popular courts of appeal. 

It is only about the former that any 
detailed information is furnished by our 
authorities. They represent the procedure 
of the ‘ judicia populi’ as consisting of two 
stages: (i.) the magistrate who means to 
impose a sentence which he knows will 
subject him to the ‘provocatio’ holds a 
preliminary investigation (anquisitio) before 
an informal assembly or contio which he 
has summoned. This lasts for three days 
and is followed by a judgment, or proposal 
as to the penalty, either in the original 
shape put forward at the beginning of the 
inquiry or in an amended form. (ii.) After 
the legal interval of three market days this 
proposal is brought by the magistrate before 
the ‘comitia’ and is either accepted or 
rejected by the assembled people. To this 
account the orthodox view—as represented 
e.g. by Mommsen, Lange and Willems— 
adds two elements: that the magistrate, at 
the close of the ‘anquisitio,’ pronounces a 
formal sentence, and that against this 
sentence a formal appeal is lodged by the 
condemned. This view, in short, represents 
procedure by appeal as having been an 
invariable element in every ‘ judicium 
populi.’ 

This theory has been so universally 
accepted since Mommsen’s review of Geib’s 
‘Criminalprozesse’ in the Neue Jenaische 
Litteraturzeitung for 1844 (p. 258) that 
no case against it has, so far as I know, 
ever been stated. Yet there are some 
difficulties in the way of accepting this 
hypothesis which are worth noting: (1) 
There is an inherent inconsistency in the 
procedure as there described. The magis- 
trate is supposed to recognize the limitations 
of his competence by resorting to a public 
‘anquisitio’ instead of toa private ‘quaestio’ 
with or without assistance of a ‘ consilium’ ; 
yet, as the result of this inquiry, he exceeds 
these limitations by pronouncing a sentence. 
Roman legal fictions have generally some 
meaning, but the fiction of a magistrate 
pronouncing a sentence which he has 
right to pronounce is 
meaningless because unnecessary, since 
the promulgation of a ‘rogatio’ would 


6 THE CLASSICAL REVIEW. 


serve the purpose required better than a 
sentence. (2) The case invariably assumed 
by modern writers, and represented by our 
ancient authorities, is that of a magistrate 
recognizing the limitations of his compe- 
tence. But let us take the case of a 
magistrate not recognizing such limitations. 
There is an appeal from his sentence, and, 
according to the current view, this would 
naturally be followed by an ‘anquisitio’ 
before a ‘contio,’ conducted perhaps by the 
same, or, after the appeal was extended to 
the provinces, probably by another, magis- 
trate. In this case we have the anomaly of 
the ‘provocatio’ being made twice against 
the same decree, except on the hypothesis 
that the ‘rogatio’ ultimately proposed 
might sometimes differ from the sentence 
originally imposed. But, even on this 
hypothesis, we have the anomaly of two 
appeals in a ‘causa capitalis, and the 
competence denied to a magistrate by 
appeal is resumed by him until a fresh 
appeal is made. Merkel remarks (Gesch. 
der Klass. Appell. p. 15) that condemnation 
by a magistrate after the ‘provocatio’ is 
wholly superfluous. If we accept this 
view, and regard the ‘anquisitio’ as a trial 
leadmg to a sentence, this part of the 
procedure should be dispensed with in a 
true case of appeal. But the ‘anquisitio’ 
could not be dispensed with. It was the 
only way in which the people informed 
themselves of the facts, and it is difficult to 
believe that a case ever came before the 
‘comitia’ without this preliminary investi- 
gation, which would not, however, be a 
trial by a magistrate. What happened in 
the hypothetical case, assumed by some 
recent writers, that the appeal might be 
made before the sentence, we cannot say. 
Merkel assumes (/.c.) that it amounted to a 
denial of jurisdiction and that no condem- 
nation by the magistrate was in this case 
possible. (3) It is often said, in support of 
the current view, that the magistrate 
becomes an accuser before the ‘ comitia’ in 
defence of his own sentence. This is true, 
but it is forgotten that he appears as 
accuser long before the question reaches 
the ‘comitia.’ The first day of the 
‘anquisitio’ is the prima accusatio (Cic. pro 
domo, 17, 45), and it is not very obvious 
how a prosecutor can pronounce a sentence. 
(4) This view of the proceedings is not 
altogether supported by the evidence of 
language, although here we are on less 
certain ground. The word anquisitio 
perhaps means an inquiry ‘on both sides,’ 
i.e. through accusation and defence (Lange, 


Rém. Alt. ii. p. 470, cf. Festus, p. 19 
‘anquirere est circum quaerere’); it 
represents a process in which the magis- 
trate and the accused produce evidence on 
either side and suggests rather the notion 
of a preliminary investigation (an évdxpucts) 
held in the interest of the assembled people 
than a formal trial which could end in 
condemnation. So too the word ‘inrogare’ 
used of the proposal resulting from the 
‘anquisitio’ is perhaps the strictly formal 
word applicable to the judgment pronounced 
on this occasion. It is a word which 
crept into ordinary use and in so doing 
changed its meaning, but its original 
signification is to ‘ propose’ not to ‘impose.’ 
It may be objected that Cicero, in the 
picture he gives of popular jurisdiction (pro 
dom. 17, 45), also uses ‘judicare’ of the 
magistrate’s proposal ; but, as two passages 
of Cicero prove (pro dom. l.c., de leg. 3, 6), 
this is simply due to the ‘usus loquendi.’ 
It was the custom to use ‘inrogare’ of 
‘multae’ and ‘judicare’ of ‘ poenae,’ prob- 
ably because ‘multae’ were by far the 
more ordinary cases in a ‘judicium populi.’ 
The strict use of ‘inrogare’ is found in Tab. 
Bant. 1. 12 and in such expressions as ‘ leges 
privatis hominibus inrogare’ (Cic. pro dom. 
17, 43). ‘Inrogare poenam’ does not 
appear to be found in Republican literature. 
For such a curious restriction of a technical 
term we may compare ‘condemnare,’ which 
was in the Empire used chiefly of ‘multae’ 
(Huschke, Multa, p.110). Nor need ‘ judi- 
care’ here mean more than ‘adjudge the 
penalty ’—the penalty, that is, embodied 
in the ‘rogatio’ which was the result of 
this inquiry. This may be all that is 
implied in such expressions as ‘ perduellionis 
se judicare dixit’ (Liv. 26, 3), or ‘perduel- 
lionem se judicare pronunciavit’ (Liv. 43, 16), 
and in these cases the word may possibly be 
used of the proposal of the magistrate before 
the ‘anquisitio.” (5) On the hypothesis 
which we are examining the capital juris- 
diction sometimes usurped by the ‘con- 
cilium plebis’ is difficult to explain. That 
a magistrate should impeach a man before 
an incompetent assembly is conceivable ; 
that a man should appeal to an incompetent 
assembly is much more difficult to under- 
stand. In the former case we have the 
misdirected power of a magistrate, in the 
latter a much less comprehensible misdi- 
rected exercise of right by a privatus. (6) 
No instance has been preserved of the 
word ‘provoco’ being used in a ‘judicium 
populi.” By itself this silence might not 
amount to much (see Mommsen on Geib, p. 


THE CLASSICAL REVIEW. 7 


248); but then Livy, in giving us the 
picture of the trial of Horatius, seems to 
treat the ‘ provocatio’ as practically extinct 
for the time of the Republic; he treats it 
as the almost vanished origin of popular 
jurisdiction, just as the author of the 
"A@nvaiwv modureta treats the appeal insti- 
tuted by Solon, which disappeared when its 
effect, the jurisdiction of the Heliaea, had 
been realized. The reason for this disap- 
pearance of the appeal must be sought in 
the change in the character of the laws 
limiting the power of the magistrate. From 
the lex Porcia of 300 B.c. these laws no 
longer permit appeal but directly limit 
competence (Liv. 10, 9). It is true that 
the law only prohibits execution and not 
sentence; but a law limiting execution is 
much more likely to grow into a limitation of 
competence than one merely permitting the 
appeal. Even laws regulating the imposi- 
tion of ‘multae’ take this form ; the multa 
suprema is one beyond which the magistrate 
may not go. The popular courts, at Rome 
as at Athens, are called into life by a 
recognized limitation of jurisdiction, and 
there was no point in keeping up the fiction 
of atrial by the magistrate. The ‘ provo- 
eatio,’ in fact, could only have appeared in 
the rare cases of actual abuse of power by 
an official. 

The chief objection to this view has been 
drawn from the words of Cicero (de leg. 3, 
12, 27) ‘omnibus magistratibus judicia 
dantur, ut esset populi potestas ad quam 
provocaretur.’ ‘Hence,’ says Mommsen 
(1.c.), ‘ the “ populi potestas,” to assert itself, 
presupposes a temporary condemnation.’ 
What Cicero does say is that all magistrates 
are recognized as judges in order to ensure 
the working of the popular courts. It is 
not denied that the ‘ provocatio’ is the 
ultimate basis for the authority of these 
courts, and jurisdiction was as necessary to 
the magistrate at Rome as it was to the 
archon at Athens to ensure a trial before 
the people. 

Nor need the lex Porcia ‘aliaeque leges— 
quibus legibus exsilium damnatis permissum 
est’ (Sallust Cat. 51) necessarily imply a 
fictitious condemnation by the magistrate in 
every popular trial. If with Mommsen we 
take ‘damnatis’ to refer to condemnation 
by a magistrate, they simply permit 
voluntary exile in cases where this condem- 
nation has been pronounced ; it is possible, 
however, that the reference is to a later lex 
Porcia which extended the theory of 
exsilium beyond the limits recognized by 
Polybius (vi. 14), in consequence of the 


employment of ‘ quaestiones extraordinariae ’ 
which excluded an appeal to the people. 

Hence it is at least possible to paint 
‘another picture of the first stage in a 
‘judicium populi.’ Criminal jurisdiction 
exercised by a sovereign assembly, in the 
absence of positive law, is largely a 
legislative act (Geib, Criminalprozesse, p- 
126) and in the ‘anquisitio’ we may see 
merely the preliminary investigation under- 
taken by a magistrate for the purpose of 
collecting and publishing evidence to 
establish the validity of a ‘rogatio’ which 
he means to formulate. 

We have, therefore, two alternatives. 
The ‘anquisitio’ is either a trial ending in 
condemnation or a mere preliminary inves- 
tigation (an integral part of the popular 
procedure) meant to furnish material for a 
‘ rogatio.’ Symmetrical reconstruction, 
which strives to connect the ‘ provocatio’ 
historically with the ‘judicia populi,’ may 
be in favour of the first; positive evidence 
and analogies rather favour the second. 
It makes no difference to the main subject 
of this investigation which of the two we 
accept, if we take this one point as certain 
—that the ‘ provocatio,’ when used against 
the magistrate’s decree, led to the ‘ anqui- 
sitio’ and was not immediately followed by 
a trial before the ‘ comitia.’ 

If this identity of procedure be assumed, 
we can answer the question whether the 
‘provocatio’ led to the reversal of a 
sentence. In the first place it is possible, 
although unlikely, that after a magistrate 
had been appealed from, he might, at the 
beginning of the ‘anquisitio, propose a 
sentence differing from that appealed from _ 
but yet subject to the ‘ provocatio’ ; it must 
have been this view of the procedure which 
led Marcel Fournier (/.c.) to speak of the 
‘provocatio’ as a change of tribunal. In 
the second place we know that the proposal 
made at the beginning of the ‘ anquisitio’ 
was sometimes amended before the end. 
Instances are furnished by the trial of 
Menenius in 474 B.o. (Liv. 2, 52, 5) when 
the tribunes ‘cum capitis anquisissent, duo 
milia aeris damnato multam dixerunt.’ [The 
word ‘damnato’ here appears at first sight 
rather a strong evidence in favour of the 
current view ; but such an expression might 
be used of any procedure where the prose- 
cution was not withdrawn (as it must have 
been where the evidence was unconvincing) 
and where, therefore, there was a case for 
the popular court; ef. Liv. 29, 21, the 
commission sent to investigate the plunder 
of Locri in 204 B.c. ‘ Pleminium—damna- 


8 THE CLASSICAL REVIEW. 


verunt atque in catenis Romam miserunt.’ 
It was not a judicial commission but one 
sent by the senate to discover whether 
there were sufficient grounds for an arrest. 
The arrest was to have been followed by a 
‘ judicium populi.’ Even in legal phraseology 
‘damnare’ is used of the prosecutor who 
effects a condemnatior (Brisonius de form. 
s. v.) ] The same procedure appears in the 
trial of Fabius in 211 B.c. (Liv. 26, 3, 7) where 
a pecuniary penalty having been proposed 
during the first two days ‘ tertio—tanta ira 
accensa est, ut capite anquirendum concio 
subclamaret.’ Whether this was usual or not 
cannot be stated. Geib regards it as very 
unusual and as contrary to the theory of 
the ‘contiones’—a view which must be 
correct if the proceedings in the ‘contio’ 
are the first stage in a legislative act. It 
was probably much more usual in the case 
of ‘multae’ than in the case of any other 
punishment ; it was easier to amend a fine 
than a capital sentence, for in some cases 
none but a capital sentence could be 
imposed, and perhaps the extension of the 
‘ provocatio’ to ‘multae’ had an important 
influence in producing the ‘reformatory’ 
procedure characteristic of a true court of 
appeal. The sentence, whether in its 
original form or amended, was then 
embodied in a ‘rggatio,’ and, after the legal 
interval required for promulgation, went 
before the ‘ comitia.’ 

No further amendment was then possible. 
Fresh promulgation would have been neces- 
sary, and for this there was no time, since 
there was not even the possibility of a 
renewal of the procedure by the same 
magistrate on the same charge (Cic. pro 
domo 17, 45; Schol. Bob. p. 337). With 
regard to the chance of amendment ‘in 
contione’ the question where the chief 
evidence was taken (a point on which we 


have no direct information) is of some 
importance. It is probable that the 
separate items of evidence were taken at 
the ‘anquisitio,’ the collected proofs being 
developed by the magistrate at the ‘quarta 
accusatio’ in the ‘ comitia.’ 

Tf, therefore, we suppose that, in the 
developed popular jurisdiction at Rome, the 
ordinary procedure of the ‘judicium populi,’ 
beginning with the ‘anquisitio,’ was 
resorted to, then the ‘provocatio’ might 
lead accidentally to a reform, not merely to 
a cassation, of the original judgment of the 
magistrate. The irregular shouts of a 
‘contio,’ perhaps also definite alternatives 
presented by the advocates of the accused 
and supported by the acclamations of the 
crowd, might lead the magistrate, during 
his own preliminary investigation, to alter 
his original judgment before its close and 
to formulate a ‘rogatio’ in accordance with 
this amended estimate of the punishment 
adequate to the offence. But, if he persisted 
in proposing a sentence which he was not 
competent to carry out and the matter was, 
therefore, forced on the ‘comitia,’ then no 
further reform of this proposal was possible. 
The ‘comitia’ could only confirm or reject, 
not amend: and it is the ‘ comitia,’ not the 
‘contio, which represents the people 
assembled for jurisdiction. 

We must conclude, therefore, that, from 
the strictly legal point of view, the ‘judi- 
cium populi, both as a court of first 
instance and as a court of appeal (if the 
two never became identified) remained a 
court of cassation, It was only accidentally, 
perhaps only occasionally, that it became a 
perfect court of appeal: but it became such 
purely by an exercise of magisterial power, 
not by an exercise of the authority of the 
court. 


A. H. J. GREENIDGE. 


NOTES ON THE TEXT OF LUCAN. 


fy the article ‘zu den handschriften des 
Lucanus’ [Newe Jahrbiicher 1893] Dr OC. 
Hosius examines in detail the readings of a 
number of passages in Lucan, setting forth 
the reasons hy which his own decisions were 
guided in his edition of 1892, That the 
general verdict goes against the Vossianus I 
Mi and in favour of the Montepessulanus 
|M] and the ‘ Pauline’ family, is a matter 
of course, Exceptions are however numerous; 


and in not a few cases the reasons assigned 
by the writer are perhaps hardly sufficient 
to let his decision be accepted as final. 

He begins with a list of passages where 
either of two variants could be supported by 
weighty references, where the weight of 
MSS tradition must determine the choice. 
He says ‘solche stellen fallen dem sieger in 
den iibrigen zu.’ That is, M and the 
Paulines carry the day. Now in testing M 





THE CLASSICAL REVIEW. 9 


our chief means, as he points out, are the 
Bernensis [B] and the Bern Scholia [C} 
where available. To these we may now 
add the Zrlangensis [E], the close connexion 
of which with B has been shown by A. 
Genthe. Of less importance for the purpose 
are the mixed MSS, U and G. From these 
witnesses we are to extract the Pauline 
reading as we best can. The correctors of 
the MSS are here noted on Hosius’ plan 
[m b vy etc.] as convenient, though not very 
precise, 

In i 687—8 quo tristis Enyo transtulit 
Emathias acies the MSS are thus reported : 
enio VC erynis MB[E]UG and a gloss ‘ dea 
belli’ in B. ‘Thus, even without E, the 
actual text of the Paulines is erinys rather 
than enyo. Hosius makes a good deal of 
the gloss. This seems to imply that B 
ought to have written enyo. But, if E has 
been rightly placed, it is strange that he 
also gives erinys: and the error, if error, 
runs through MUG. To quote passages 
from other writers in which enyo occurs is 
beside the mark: nobody doubts that it is 
possible, though the name does not come 
elsewhere in Lucan. And to follow this up 
by suggesting that the scribes who wrote 
erinys had the tristis erinys of Verg. Aen. il 
337 and other passages running in their 
heads makes this process of reasoning quite 
grotesque. Not only B[E], but M, the 
value of which is so largely based on the 
sober ignorance of its writer, are guilty. 
If we throw the charge back upon the 
archetype of these ‘ Paulines’ we shall have 
to assume that the Scholiast C in his lemna 
deliberately corrected an error [or made a 
different choice out of two variants] without 
adding a word of comment, or reporting a 


variation. Such an assumption is quite 
unnecessary, and does not rise above 
possibility. If the guilt rests on Mb[E] 


themselves, their agreement is astounding. 
As to the gloss ‘ dea belli, it is well known 
that inference from such comments is a 
risky venture. We have here first to be 
sure that ‘dea belli’ cannot possibly be a 
comment on erinys. In iv 187—8 iam iam 
civilis EHrinys concidet the Fury that 
promotes civil war is hardly distinguishable 
from a war-goddess, indeed she is such for 
the purpose in hand. When we turn to 
the scholia on other passages we find 
notable things. On viii 90 me pronuba 
ducit ELrinys C quotes Vergil’s et Bellona 
manet te pronuba, and Weber gives a similar 
note from Schol. Berol. 3 and Voss. In x 
59 Cleopatra is Latii feralis Lrinys, and on 
63 terrwit ete. C says ‘mentionem facit belli 


Actiaci,’ while on 59 Schol. Berol. 5 roundly 
gives ‘ Bellona’ as an explanation of Hrinys. 
I now venture to say that I set no store by 
this gloss, that the argument from it to 
Enyo is in my view fallacious, and that the 
Pauline text Hrinys has not to my mind 
been successfully assailed. In fact I would 
let the passage ‘fallen dem sieger in den 
iibrigen zu,’ which is Dr Hosius’ own rule. 

Of the rest of the passages referred to in 
this list it is perhaps unnecesary to speak 
in detail. But it may be pointed out that 
the principle assumed throughout the 
paragraph is as follows. When the text of 
a& passage in an author is disputed, the 
probability of this or that reading in itself, 
apart from weight of MSS, is increased by 
the occurrence of similar expressions in 
other authors or in other parts of the same 
author. Thus in iv 304 dwris silicum lassata 
metallis |medullis| witnesses can be cited in 
support of either reading. But medullis is 
only the original reading of V, afterwards 
altered [V1], and is rejected on the ground 
of insufticient MS authority. So far well. 
But when we go on to say that the interpo- 
lations of V are sometimes very apt, that 
they are often countenanced by other 
passages in Lucan, of which they may be 
reminiscences, we find ourselves on very 
different ground. Thus in ix 446 nullasque 
timens teliwre [sentire WV] procellas we 
suspect that vi 470 sentire procellas was the 
cause of the corruption. In ii 133 hominis 
quid fata paterent [| pararent V| we may 
have an echo of 11 68 parabant and vi 783 
quid fata pararent. We may with Dr 
Hosius take a further step and find echoes 
of other authors. Thus viii 539 qua tellus 
Casiis excurrit [exultat V1] harenis may be 
traced to Aen. ii 557. 

The worst of it is that the argument does 
not apply to V only. We have seen 
above that tristis Erinys has been regarded 
as perhaps an echo of Aen. ii 337 ete. Here 
the Pauline MSS are implicated. In vi 200 
where limine portae is the MSS reading, 
save that VB and possibly C have torta [the 
variant is given in E], we are told ‘/imine 
portae ist ein beliebter versschlusz,’ and 
references are given. If we have here a 
corruption and its origin, then all the MSS 
[C included] are wrong in /imine and 
MUG[E] in portae. It may be so, but I 
wish I could believe that Hosius’ ba/listaqgue 
limite torta is right. He does not translate 
the words, and of his two quotations one 
has limes datus, the other limitem agit. 
These do not establish Jimite standing by 
itself, and promoveat = summoveat is also in 


10 THE CLASSICAL REVIEW. 


need of support. I do not see my way here 
at all. 

That V is far more open to the suspicion 
of conscious interpolation than M and the 
other Paulines is, I think, fully made out. 
But, as Hosius remarks, the recension 
represented by V is very old. Priscian 
cites iv 131 with robore [V'] instead of 
vimine. This brings us to about. a.p 500. 
If the palimpsest fragments N are rightly 
assigned to the 4th century, we may perhaps 
set it back further, for in v 375 we have 
retinere [VN] for revocare. And we have it 
on record that the text was corrupted very 
soon after Lucan’s death. There is therefore 
need of the greatest caution in the use of 
references when we are trying to make 
out which is the genuine reading of a 
passage and which the corrupt. Thus in ii 
553 Seythicis Crassus victor remeasset ab oris 
[so MGC, arvis VUBE] it is surely a bold 
assertion to say that Valerius Flaccus iv 
589 remeat qui victor ab oris Bebryctis had 
this place in view and oris in his text. Nor 
does it serve much purpose to prove from 
other writers that Scythicae orae is a 
common combination. Nor that victor ab 
oris is found in Statius, remearat ad oras in 
Silus. Lucan is thinking-of the plains of 
Parthia, and we might as well point out 
that we have arva used in cognate passages 
iv 743 [Africa], viii 370 [Mesopotamia], ix 
626 [Libya], 697, 749, 939 [the same]. Yet 
oris is surely better as giving the general 
sense of ‘ borders’ ‘realms’ ‘ parts,’ and it 
has good authority, particularly M, in its 
favour. No amount of references will help 
it, or make it a case to support’ the 
preeminence of M. For with what face can 
we argue in vi 200 that limine portae is a 
favourite ending and so discredit it, while 
here we argue that oris is right because it 
so often comes in similar expressions ? 

Again in ii 609 Brundisii tutas concessit 
Magnus in arces [conscendit VG] we may 
admit that concessit is right. But the 
references only prove usage: that is, prove 
that concessit will stand. Stat. silv. iii 3 163 


takably appears. 


Diomedeas concedere wussus im arces 1s 
probably an echo of the passage, I grant. 
But, if MBUC[Ej had conscendit and VG 
concessit, Should we be justified in adopting 
the latter? I doubt it. For wecan hardly 
prove a negative against conscendit. In 
fact it is the MSS authority, far more than 
the references, that determines our choice 
in such passages. And this authority must 
be established without depending on such 
passages, or we argue in a circle. 

There are plenty of places where the 
superiority of M and the Paulines unmis- 
Such are v 189 tam 
magna [multa V], 210 locutae [locuta est 
VU], 696 ad fatum belli [ad summam VU], 
vi 330 condixit iter [convertit VG], 496 
valuere [potuere VU], 508 nimiae pietatis 
[minimae V}, and many more. 

That V on the other hand has the advan- 
tage in some places is also to be admitted. 
From the instances given by Hosius at the 
end of his article I select ii 303 prosequar 
VGE [persequar MBUC], 614 linguam 
[swlewm MB!K, possibly the representative 
of a lost word], iv 329 tamen [siti MBE, 
very like a gloss], ix 795 pollente [tollente 
MBEu]. But others seem to me doubtful, 
as vi 112 foliis [morsu or morsus MBE], vii 
641 vineimur |vincitur MBEUG], ix 165 
saevus |saevas MBEUG], 760 cruore [veneno 
MU]. And inv 107 notas [totas MBEUGC} 
I greatly prefer totas. Some others I omit 
as of little or no importance. And it will 
be observed that in many of the above 
passages V does not stand alone. 

So much for the matter in general and 
the use of references in particular. In 
another paper I hope to comment on some 
of the passages treated by Dr Hosius in 
detail, possibly on others not referred to by 
him in the above article. I think also that 
there is still something to be said on the 
subject of the lines omitted in some of the 
MSS, though I agree with the editor that 
certainty can hardly be attained. 


W. E. HEITLanp. 


MILTON AND PINDAR. 


In the October number of the Classical 
Review Prof. Campbell quotes the following 
passage from Milton’s Sonnet on ‘his three 
and twentieth year’ ;—~ 


Yet be it less or more or soon or slow, 

It shall be still in strictest measure even 

To that same lot however mean or high, 

Towards which time leads me and the will 
of heaven, 





THE CLASSICAL REVIEW. 11 


in illustration of Pindar Nemean iv. 37 :— 


> ‘ xe ¢ > ‘ 
€/L0l é OTOLaV ApeTaV 
™ ‘ »” 
€dwke TOTILOS avaé 
> 9) 2 , ” , ‘ 
ev old Gru xpovos EpTwv TeTpwyevav TEeAETEL. 


Prof. Campbell observes that ‘literary 
parallels appear differently to different 
minds.’ Availing myself of this concession, 
I venture to suggest that the parallel in 
question is more apparent than real. ‘Torn 
from their context, it is true, Pindar’s 
verses, which are among the most inter- 
esting he ever wrote, approximate closely 
to those of Milton. But if they are read in 
connexion with what immediately precedes 
and follows them, it will, I think, appear 
that the attitudes of the two poets are 
widely different. Milton’s lines, as I 
understand them, are an expression almost 
of humility, certainly of resignation to the 
divine will, ‘however mean or high’ his 
future lot may be. But the tone of Pindar 
is just the reverse: he has no doubt at all 
of the lot to which ‘King Fate’ is leading 
him. In many of his Odes, but nowhere 
more clearly than in this fourth Nemean, 
Pindar announces important literary inno- 
vations which he has introduced into the 
substance of the éruikwov—iunovations 
which, as from this and many other 
passages is abundantly proved, had 
provoked the bitterest hostility from 
contemporary critics, the ddio. of 1. 38. In 
spite of this hostile criticism he is deter- 
mined to persevere— 


eEvauve, yAveia, kal 703° adrtixa, poppryé, 
péedos— 


and confidently predicts that he will at last 
weather the ‘sea of troubles’ that holds 
him by the middle and reach the terra firma 
of a glorious immortality : 


oddpa d0fopev 
daiwv iméprepor ev pac xataBatvev. 


Pindar’s érwikia have survived, while 


those of his rivals have perished, and the 


principles which his dperd tanght him to 
impose upon the Ode have been accepted by 
succeeding writers of every language. His 
position then in Aegina or Syracuse about 
470 B.c. very closely resembles that of 
Wordsworth in England on May 21, 1809. 
Wordsworth also had his poetic innovations, 
which had provoked a no less hostile 
criticism, and the genius of the English 
poet is as confident as that of the Greek, 


In his well-known letter to Lady Beaumont 
(Memorials of Coleorton ii. p. 9), referring 
to the unfavourable reception of his poems, 
Wordsworth uses prose which exactly repro- 
duces the spirit of Pindar’s verse: ‘Trouble 
uot yourself upon their present reception : of 
what moment is that compared with what I 
trust in their destiny }’ And Wordsworth’s 
description of ‘ what is called the public’ in 
this famous letter is an admirable comment 
on Pindar’s picture of the same class in his 
own day: 


Kpayérat d€ KoAoLot Tarewa venovtar (V. 3, 82). 


These boasts of conscious genius, so 
literally fulfilled, have so peculiar an 
interest that, at the risk of appearing 
prolix, I shall venture to quote one example 
more. Goldsmith, in early manhood, 
slighted or unknown, thus writes from his 
garret to a college friend: ‘God's curse, 
Sir! who am I? Eh! what am I? Do 
you know whom you have offended ? A man 
whose character may one of these days be 
mentioned with profound respect in a 
German comment or a Dutch Dictionary, 
There will come a day, no doubt it will—I 
beg you may live a couple of hundred years 
to see the day—when the Scaligers and 
Daciers will vindicate my character, give 
learned editions of my labours, and bless 
the times with copious comments on the 
text.’ Goldsmith is now a ‘decolor Flaccus 
nigerque Maro’ in Germany: but how 
literally this prediction, too, has been 
fulfilled will still better appear from the 
following translated extract from a letter of 
Goethe to Zelter :— 

‘I lately chanced to fall in with the 
Vicar of Wakefield, and felt compelled to 
read the little book over again, from 
beginning to end, being not a little affected 
by the vivid recollection of all that I have 
owed to the author for the last seventy 
years. The influence that Goldsmith... 
exercised upon me, just at the chief point of 
my development, cannot be overestimated. 
This high benevolent irony, this just and 
comprehensive way of viewing things, this 
gentleness under all opposition, this 
equanimity under every change, and 
whatever else the kindred virtues may be 
termed—such things were a most admirable 
training for me.’ 





W. T. Lenprem. 


Pror. CAMPBELL points out in the October 
number of the Classica/ Heview a passage in 


12 THE CLASSICAL REVIEW. 


Pindar’s Wemean Odes (iv. 67), parallel to 
onein Milton’s Sonnets. Perhaps as clearly 
parallel to 


‘Toward which Time leads me and the will 
of Heaven’ 


is the expression edOvropros aidy in NV. ii Da 
But even stronger evidence that Milton was 
a student of Pindar is to be found in the 
words ‘pearléd wrists’ used in reference to 
the Nereids in the Comus. Surely this far- 
fetched epithet rests on a misapprehension 
(shared however by Paley and possibly by 


early commentators) of the meaning of ayAa0- 


xaprov Nypéos Ovyarpa in N. iii. 56. 
Again, in Jsthm. iii. (iv.) 41 the words 


2 4 \ / / 
€v UTVM yap TETEV (papa), 
> ye , lal , 
GAN avayerpopeva. xpoTa Aaprret, 
> / \ a 4 > + 
Awaddpos Oantos ds dotpo.s ev adAo1s, 


may well have supplied the key-note to the 
sublime passage in Lycidas, 


So sinks the day-star in the ocean-bed, 

And yet anon repairs his drooping head, 

‘And tricks his beams, and with new-spangled 
ore 

Flames in the forehead of the morning sky. 


Lastly, ‘the dear might of him that 
walked the waves’ might have arisen from 
the depois dvaykas Xepot of JN. viii. 3; and 
‘build the lofty rhyme,’ though itself quite 
in the Miltonic vein, might however have 
been suggested by P. iii. 115 


Lal , e LN 
e€ éréwy KeAadevvv, TEKTOVES OLA woot 
dphoocav. 


R. Y. TYRRELL. 





DESCRIPTIVE NAMES OF 


In his interesting article on ‘ Descriptive 
Animal Names in Greece’ (Class. Kev. for 
Nov. 1894), Mr. A. B. Cook calls attention 
to the error of the view which attributes a 
certain type of phraseology to an oracular 
or religious dialect, and points out that it 
is more probably toa large extent provincial 
in origin. From his arguments on this 
point few will dissent ; but the speculations 
which succeed them touch on a slippery 
subject, and as there is no argument so 
dangerous as a false analogy, unless it be a 
partially true one, some criticism of his 
main point may not be out of place. It 
seems to me that there is very little ground 
for the comparison between these and 
similar names on the one hand, and on the 
other the euphemistic appellations applied 
to various animals by the savages of the 
present day, to whom he refers in such 
detail. It is obvious that almost every one 
of the animals which are indicated by these 
‘deferential periphrases’ is an animal 
which it is dangerous or unlucky to offend. 
On the other hand it can hardly be said 
that the ancient Greek descriptive terms 
indicate fear of this kind. Such animals as 
the lion, tiger, snake, bear and wolf are 
most prominent among those for which 
savages invent nicknames; the lightning 
again is clearly a power which should be 
spoken of with respect ; and a superstitious 
race will naturally speak circumspectly of 


ANIMALS IN GREECE. 


things on which their life depends, or of 
which they happen for the moment to have 
need. But the animals whose picturesque 
Greek. names are adduced belong as a whole 
to quite a different category. These names 
in fact simply indicate some striking cha- 
racteristic. They are exactly parallel to 
names of plants founded on some peculiarity 
of colour or shape. If one part of the animal 
or plant is particularly striking, that is 
emphasized in the name; if not, the 
prevailing effect finds its expression. Thus 
with the words given by Mr. Cook, to 
which may be added, e.g., dacvrous (hare) 
and Bpayvxédados (a kind of fish), I would 
class all such names as dacvAAis (bear), 
ovvddous, Euias, aiodias, alburnus, épvOpivos 
(kinds of fish), xvavos (a bird, the wall- 
creeper), wop@upiwy (purple coot), épvbpdvor, 
épvoicxymtpov, albucus, albuelis, (plants), 
and Mdxpwv (the name of a Pontic people). 

These names then are ‘to be attributed 
merely to the inborn poetry of rustic wits,’ 
and to nothing of the kind which has given 
names to dangerous creatures. Both have 
their origin, it is true, in the savage or 
rustic imagination, but in the latter case 
a imagination has been stimulated by 
ear. 

With regard to Mr. Cook’s explanation 
of the name Melampus, some naturalist 
ought to be consulted as to whether the 
name ‘black-foot’ is so ‘peculiarly appro- 


THE CLASSICAL REVIEW. 13 


priate to a goat.’ So far as I know, the 
feet of the common goat may be of any 
colour proper to bone. The hoofs of the 
ibex, though dark, are not black, or at any 
rate not much blacker than the rest of the 
leg. The name ‘ black-foot’ would hardly be 
applied to any creature whose feet were not 
strikingly differentiated from the body by 
blackness. Though there may be nothing 
to say against the goat-origin of Melampus 
on the score of the other evidence adduced, 
it certainly seems to me that the evidence 
of the name itself does not help to that 
conclusion. 
G. F. Hii. 

Tue interesting paper by Mr. A. Bernard 
Cook in the November number of this 
Journal, and especially the list of para- 
phrases for animals at p. 382, have sug- 
gested to me an explanation of a passage in 
the Homeric Hymn to Demeter which 
hitherto has been unintelligible. 

The lines in question are vv. 227 sqq. and 
run as follows : 


ov pv Eola Kakoppaddyat TOHVYS 
ott’ ap’ éexndvoin SnAjoetat ov6’ trotapvov' 
olda yap dvtitopov peya péptepov bAoTOpoL0, 
olda 0’ éexnAvoins ToAvTiLoOvos eo OAOv EepvTpov. 


Demeter, disguised, accepts the charge of 
Metaneira’s child and says ‘it will not be 
from his nurse’s folly that sorcery or izo- 
tapvov hurt him; for I know an antidote 
much stronger than tAordopos(ov), and I 
know a good defence against sorcery.’ 
*ExyAvoty is a familiar word, but what are 


bAoropoww and tzorayvov! Ruhnken, who 
first edited the Hymn, said ‘ab hoc ulcere 
manum abstineo ;’ later editors have been 
more venturesome, but without gain to the 
passage. Neither word will yield to emend. 
ation ; tAorduo.o in particular wears an air 
of solidity, and is defended by the jingle 
with dyvriréuowo. I propose therefore to 
treat it as an epithet on the analogy of 
depeotxos = snail, dvooreos = cuttle, (pis = 
ant, all Hesiodean. What animal or insect, 
suitable to the context, could be figuratively 
described as tAorépost Perhaps the worm, 
in prose language é€Apws or oxwAné; and 
Demeter will be promising to save Metan- 
eira’s infant from a familiar scourge. Ovid, 
Fasti iv. 358 sq. quoted by Gemoll, repre- 


sents the child as already ill, but he does 


not specify the malady. It is obvious that 
both the speaker and the context are suit- 
able to the superstition of ‘avoiding the 
risk of offending the animal by using some 
periphrasis of a deferential sort in lieu of 
the actual name.’ ‘Yzorapvev will become 
simply the present participle of trordyvw 
(the scribe’s accentuation is naturally imma- 
terial) and mean the ‘eater,’ or ‘borer.’ 
The worm, I see from Mr. Cook, was called 
by Syracusans yadayas, and we find in Hesy- 
chius tAoujrpa: eldos oKwAnKos. 

The translation will then run: ‘neither 
sorcery nor Borer shall hurt him, for I know 
an antidote much stronger than Woodeutter 
etc.’ If this account be correct, it is an 
instance of how the meaning of these docu- 
ments is to be sought by explanation rather 
than by free mangling. 

T. W. ALLEN. 


NOTES ON EURIPIDES’ PHOLNISSAE. 


‘Tue recent publication of Professor Weck- 
lein’s valuable edition of Euripides’ Phoen- 
issae (Leipsic, 1894) has prompted me to 
put forth certain conjectural emendations 
upon the text of that play. For the sake 
of perspicuity and brevity I place the 
reading I would suggest at the head of each 
of the following notes :— 

Vv. 208--213. 
“loviov Kata wovTov éXa- 
Tat TAEvoara TEpIPpLTWV 
imtp dxaprictwy Tedlwv 
évariars Lepipov rvoais, 
ot mvevoavTos ey otpavar 
KdAAorov KeAddnpa. 


‘Having sailed down the Ionian sea in a 
ship, over the watery (zepippvtwv) uwnhar- 
vested plains, by the sea-breaths of Zephyr 
whose breath in the sky causes fairest 
melody.’—V. 211 SuxeAéas Wecklein with 
the MSS. V. 212 inmevoavros Wecklein 
with ‘the MSS.—The emendation évaAcats 
(which may be supported by /e/. 1459 sq. 
kata pev ioria mwetacar’ at-|pars AurdvTes 
évadiais: We must remember too the Homeric 
expression that Euripides seems to have 
had in mind, dxpaj Zédvpov xeAddovr’ eri 
olvora mévrov) helps to get rid of that 
troublesome circumnavigation of the Pelop- 
onnese. Did 'Idvov rovrov help to bring in 


14 THE CLASSICAL REVIEW. 


Sixedias? We naturally think of Hddov in 
Bacch. 406.—The emendation ot zvevoavros 
makes the aorist participle intelligible in 
an ingressive sense. Wecklein’s comma 
after immevoavros and change of ovpavar to 
dppévor do not satisfy. 


Vv. 473—477. 

éyo dé—matpos yap dopwy mpovorKeyapenv 
Tobpov TE Kat TODD —expuyely xpyiLov pas, 
ds Oldirous EbOeyEar’ cis 740s TOTE, 
e&qdOov eEw THAD Exov avTos xGovos, 
Sods TLD dvdcoew TaTploos eviavTOD KUKAOV, KTE. 

Professor Wecklein reads dopdrwy with 
the MSS. in 473 and follows Hartung and 
Paley in bracketing v. 476. The awkward. 
ness of this is patent.—The corruption of 
vip Sduwv to dwndrwv seems due (in part at 
least) to dappdxov in v. 472. 


V. 504. 
dotpwv av Ooi’ 7)6€us Tpos avToAds KTE. 

The dotpov dv Gow’ HAiov of the MSS. 
is changed by Professor Wecklein to dvw 7 
dv Oo’ yAtcov—a desperate guess.—The 
corruption 7Aéov seems chiefly due to A read 
as A. 

V. 703 sq. 
nkovoe peiLov avTov eis 144s ppovery, 
kydec Tt “Adpaatov Kat otpara. reroubora. 

V. 703 7 OnBas (for cis yas) MSS., 7 
6vnrov Wecklein, after Kinkel. Professor 
Wecklein had also thought of cis O@7Bas.— 
eis ypas (for which a partial support is to 
be found in Hipp. 6 dco dpovotow eis Hpas 
péya) is palaeographically possible enough, 
a combination of uncial and minuscule 
blunders readily producing 7 OyBas. The 
comparative too played its part. 


V. 740. 
rl Onta SpOpev; Gropia yap, ei pevo. 
The dzopiav yap ob pevd of the MSS. and 
Wecklein is certainly much less effective. 
The corruption is of a familiar type. 


V. 747. 
ippotepov’ adm@oAnPOev yap ovdey Oarepov. 
‘Both; for either taken by itself is 
nothing.’—dydotepov' dmrodapbev yap ovdev 
Garepov MSS., apdorep’* ev drodapbev yap 
ovdev Garépov Wecklein. 


Vv. 881--8853. 
woAXot b€ veKpol Tepl VeKpols TETTWKOTES, 
"Apycta kal Kadpeta p<e>igarvtes pp én, 
muKpovs yous dwacover OnBalar xOovi. 
In v. 882 Professor Wecklein reads with 
the MSS. pigavres BeAn.—The correction pérn 
has already been suggested in the Critical 


tais 8 éBSdpats "Adpactos ev TiAaow HV, 


Appendix to my edition of the <A/cestis (on 
vy. 304). 
V. 947. 

obtos 8¢ waAOS THU Gvypmwevos TOAEL KTE. 

Professor Wecklein, with the MSS., 
dveysévos (‘ hingegeben, dargebracht ’). Does 
not ‘attached to’ seem more naturalin view 
of the context ? 


Vv. 11384—1138. 
The following transposition (with one 
slight emendation) I vénture to offer as a 
possible solution of a difficulty. 


1134 
Udpas éxwv Aaotow év Ppaxtoow 1136 
’Apyciov avynp’ dorid éxxAnpodv ypadye 
; 1137 and 1135 
Exatov €xLOvav" ek O€ TELXEwv LéETwV 
1135 and 1137 
Spaxovres Edepov téxva Kadpetwv yvabors. 1138 
V. 1135 éxaAnpdv MSS. and Wecklein, 
éxaAnpoov Geel. The readings éyidvais (under 
EBddpats) and éxzAnpov might well be due 
to the position of v. 1135.—Professor 
Wecklein keeps the MSS. order but brackets 
ypadye and éxywv Aaotow év Bpaxioow. ‘ Die 
eingeschlossenen Worte,’ he writes, ‘ welche 
die Konstruktion stéren, scheinen inter- 
poliert zu sein.’ 


Vat 93: 
” 2¢/ > , + e 
eOpwroKkov e&erimtov avtvywv amo, KTE 
eOvytokov, e€emittov Wecklein with the MSS. 


V. 1233 sq. 

ipeis 6 aydv apevtes oikelav ybova 

vicgecOe KTE. 

The best MSS. read ’Apyeto. yOova: two 
inferior MSS., ’Apyefav x@ova:! Professor 
Wecklein, adopting a conjecture of J. Weid- 
gen, reads ’Apyeio., taAw (for his opinion 
about the last word in vs. 1232 see the 
Appendix.)—In support of my conjecture I 
would cite Soph. Ant. 1203. ’Apyetor seems 
due to v. 1238, just as in Eur. 7.7. 588 
ayyetdat is due to the occurrence of the same 
word in the same place in v. 582. (’Apyetav 
may well be, as Kirchhoff thought, a late 
correction, though possibly a gloss on 
oikelay.) 

Mortimer Lamson Earte. 


' ‘apyelav, quod est Ald. et recentiorum, videtur 
etiam esse in be, ea correctura mihi videtur mani- 
festa,’ writes Kirchhoff. The MS. c=Laurentianus 
(Kirch. Florentinus) 31, 10. Von Wilamowitz- 
Moellendorff (who designates it as O) gives a good 
account of this MS. in the case of the Hippolytus 
(see his edition of that play, p. 181). 1 do not 
know Serer Kirehhofi’s ‘videtur’ has been 
verified, 


LL 


THE CLASSICAL REVIEW. 15 


NOTE ON SOPH. 


arias 8 irép perdbpwv dhovdcaw dudtyavov 
Kiko 

Adyxats Exrarvdoy cropa 

éBa xré. 


Tue fact that the army-eagle (for only so 
can one represent the interlocking of sign 
and thing signified in this splendid passage) 
is depicted ‘agape with blood-thirsty spears 
about the seven-gated mouth’ seems quite 
enough to warrant some attempt at emend- 
ation. But the simple and handy correction 
of oropa to wodAw (Blaydes) or zodwp’ 
(Nauck) does not explain at all how oropa 
came into the text. The conflict of 
dudixyavov and ordua in the vulgate suggests 
their reconciliation—ap¢iyavov ordpa, ‘ with 
mouth agape’; but then we must change 
éxtamvAov to éxrarvAwt,—and this is 
precisely Semitélos’s inevitable and admir- 
able correction. But he has not quite 
finished the good work ; for we observe that 
in the strophe we have érrawvAw near the 
beginning but not in quite the same place 
as in the antistrophe. According to the 
principle so largely followed by the 
tragedians we might expect exact corre- 
spondence in this regard between strophe 
and antistrophe here. In the  strophe 
érramvAwt is evidently in the right place; 
for it cannot be moved to correspond with 
érrarvAwt in the antistrophe without 


ANTIG, 117—120. 


spoiling the verse. but in the antistrophe 
éxramvAw and dudiyavev can change places 
without affecting the metre, and by making 
them shift their positions we bring together 
elements that belong together in sense— 
ertamvAw. KikAwe and dudiyavev cropa. We 
thus see that the corruption of érrarvAwc to 
éxtamvAov is due to its false collocation with 
oropa, the word xixAor thrown together 
With dpdtxavev being, not unnaturally, 
taken as an adverb repeating dude J 
would, therefore, read and point thus (the 
pointing agrees with Professor Jebb’s) :— 





otras 8 irép perdbpwv, dovoraow exraridur 
KUKAwL 

Aoyxais dudiyavov cropa, 

€Ba, K7e. 


I would add that the thing signified is 
obviously the van (ordpa) of the army 
bristling with spears. 


In Antigone 1 it seems not to have been 
observed by those that suspect (Nauck) or 
would emend (Wecklein, M. Schmidt) the 
word xowov, that Sophocles had in mind 
when writing this verse Aesch. Prom. 613 
® Kowov @deAnua Ovytoiaw daveis—the 
metrical equivalent, syllable for syllable 
and caesura for caesura, of Ant. 1. 

Mortimer Lamson Earte. 


ENCLITIC W#. 


CONSIDERATION of various notes on -ne as 
particula confirmativa leads me to hope 
that a classified list may still be useful of 
passages in which -ve is not used as a parti- 
cle of interrogation, or in which it is 
appended to relatives. Setting aside -ne in 
the principal clauses of questions, and also 
the elliptical -ne ut or utne, I have collected 
the following instauces from Minto Warren’s 
paper (Amer. Jowrn. Phil. vol. ii.) and from 
various commentators. 

(1) -ne strengthens verbs, Z7’rin. 129, 136 
(Brix: but it is better to regard these as 
interrogative) and Jost. 580 


*Reddeturne igitur faenus? , Reddeturne ; 
abi.’ 


So Sonnenschein with Leo: reddetur nunc 
abi MSS. reddet : nunc abi, Schoell. 

Leo’s reading derives some support from 
the next group. 

(2) -ne is used in questions and answers 
of the form Egone? ‘tune. So Capt. 857, 
4, 2,77; Epid. 576, 4, 2,6; Trin. 634, 3, 
2, 8. Similarly itane Pers. 220, 2, 2, 38, 
while Stich. 635, 4, 2, 52 we have egone? 
Tune! Mihini? Tibine! Most. 955, 4, 2, 39 
Schoell with the MSS. reads Egone {—Tu— 
Tu né molestu’s, but Sonnenschein Egone !— 
Tune—Tun molestu’s, 

3) In assertions we have hicine Jost. 
508, 2, 2, 76 (hicine pereussit !) ; Zpid. 541, 
4, 1, 14; tun Most. 955 (Sonnenschein) : 
Schoell makes Most. 508 a question. 


16 THE CLASSICAL REVIEW. 


The MSS. give nonne Phorm. 969 Nonne 
hercle ex re sues me instigasti, Demipho : so 
in oratio obliqua, Merc. Prol. 62  sese, 
Nonne, ut ego, amori operam dedisse. 

W.B.—Perhaps in Andr. 683, 4, 1, 59 
hem! nuncin demum! we should under- 
stand restituis? a virtual command like 
abin? Most. 850 or manesne 887a. 

(4) We have three instances of men, or 
ten forming a group by themselves. 

Pseud. 371, 1, 3, 1387 
A. Numquid aliud etiam voltis dicere!— 

B. Ecquid te pudet ? 


A. Ten amatorem esse inventum inanem 


quasi cassam nucem. 
The acc. and infin. seem to be dependent 
on pudet. 

Eun. 931, 5, 4, 9 

nam ut mittam, quod . 

tum hoc alterum 

(Id verost quod ego mihi puto palmarium) 

Men repperisse. 

Ennius ap. Cic. de Div. i. 
dolet men obesse. 

This group and the two instances of nonne 
seem to show that sentences of the form 
‘ Mene incepto desistere victam !’ are origin- 
ally purely exclamatory. 


31, 66 hoc 


(5) hicine, hocine, haecine (n. pl.), egone, 
tune, tibine (?) are found in protasis (sé, 
once whi) and (%) in apodosis. The word 
emphasized by -ne is the first in the sentence 
except Heaut. 950, 5, 1, 77 Sed Syrum— 
Quid eum !—egone si vivo, adeo exornatum 
dabo (egone MSS. : ego Umpf.) ; in pretest: 
Hd. Td, 1, 1, 03 ; Will. 309,°2) 3,38: 
565, 2, 6, 82: 936, 3, 3, 62 (egone hoe si 
Speuinm .sl1_ hodie Gene dolum dolamus, 
Quid tibi ego mittam muneris) ; Adelph. 770, 
5, 1,8. It might be doubted whether -ne is 
in the apodosis or protasis in Andr. 478, 3, 
1, 20; Heaut. 950, 5, 1, 77, but there is 
only one instance in which -ne could not be 
in the subordinate clause, Curc. 139, 1, 2, 
51 tibine ego si fidem servas,...statuam : 
and here perhaps we have not tibine but 
n(e) ego. If so -ne probably belongs to the 
subordinate clause in 


(6) Asin. 884, 5, 2, 34 egone ut non 
subripiam...Non edepol conduci possum, to 
which Poen. 421, 1, 3, 19 is similar. 

(7) -ne is attached in complete sentences 
to relatives. 

(a) -ne merely strengthens, 

Cist. 653, 4, 1, 2 
Nullam ego me vidisse credo magis anum 

excruciabilem 
Quam illaec est, quae dudum fassast mihi, 

quaene infitias eat. 


If quae is really relative here, it is difficult 
to extract an ieee rogative meaning. 

Truc. 534, 2, 6, 52 
Is (se. ills) ‘te dono—Poenitetne te quot 

ancillas alam 
Quine etiam insuper deducas, 

comedint cibum 4 
So Schoell. Quin etiam men super adducas 
MSS. 

‘Quin’ tuetur Dombart, 
p- 736, coll. 228, Amph. 235, Asin. 419, 
Cure. 209, quia insit in v. ‘ poenitet’ 
negandi vis : Nonne satis multas ance. alo, 
quin plures me alere cogas?‘...Si quid 
mutandum “ Qui mihi” scripserim (Ussing). 
The passage seems rather to mean ‘ Do you 
think I have too few slaves, seeing that you 
add, &ec.’: quin...adducis! would suit the 
petulant tone of. the epee 

Here add Hor. Sat. 1 21 (O seri studio- 
rum quine putetis), if ins qui be relative. 

In Rud. 538, 2, 6, 53, Ter. Ad. 261 (cited 
by Palmer), quin is an emendation that is 
now abandoned. 

(8) -ne strengthens the interrogative 
force of a question. The relative is the 
first word in the sentence: the principal 
clause may itself be introduced by a word 
of interrogation. 

Stich. 501, 3, 2, 45 

quaen eapse deciens in die mutat locum 

eam auspicavi ego in re capitali mea ? 

Mil. Gl. 614, 3, 1, 20 

Quodne vobis placeat displiceat mihi ? 
(-ne B, D, F, Ritschl: nec C) 

Rud. 272, 1, 5, 14 
Quaene ejectae e mari simus ambae obsecro 
Unde nos hostias voluisti hue adigere ? 
so Fleckeisen: but Sonnenschein puts ques- 
tion at obsecro. 

Cist. 675, 4, 2, 7 

Quamne in manibus tenui.. 

Ubi sit nescio. 
These examples suggest that -ne belong to 
the subordinate clause, and not to the 
governing verb that is understood, in con- 
structions like J/ost. 556, 3, 1, 29, quid 
censes!—egon quid censeam? (egon MSS., 
Schoell, Sonnenschein), or Hor. Sat. 2, 5, 
18, Utne tegam. 

(8) -ne is attached to relatives in incom- 
plete sentences, giving to the relative clause 
in (?) all cases an interrogative force. 

The relative clause relates to a_ state- 
ment just made by another: pid. 717, 5, 
2,52; Merc. 573 (565), 3, 3,12; Most. 738 
(724), 3, 2,49; Rud. 767; Andr. 768, 4, 4, 
29. To these is akin Rud. 1231, 4, 7, 5 
A. Aequon videtur tibi, ut ego alienum 

quod est 


quae mihi 


Philol. xxviii. 


.cistellam 


THE CLASSICAL REVIEW. 


Meum esse dicam?—B. Quodne ego 
inveni in mari } 

There is an assertion involved in alienum 

quod est, to which the quodne clause objects. 

Sometimes the -ne clause follows upon the 
demand of another, Amph. 697, 2, 2, 65; 
Cure. 701, 5, 3, 27 ; Rud. 272 (according to 
Sonnenschein’s stopping), 7b. 1019, 4, 3, 80; 
Trin. 360, 2,2, 79; Phorm. 923, 5, 8, 30. 

In Mil. Gl. 972, 4, 1, 27 quae is reading 
of MSS. and Goetz. Jb. 13 Quemne ego 
servavi? can hardly be an instance of -ne 
added to the relative, for there is no possible 
antecedent. Rather quem is interrogative, 
and we must compare the Horatian quone, 
quantane. In all these instances save (1) 
Mil. Gl. 13, and Rud. 767 (A. ignem mag- 
num hie faciam. B. Quin inhumanum 
exuras tibi?), the relative clause expresses 
astonishment, or conveys an objection or 
protest against what has just been said. In 
one case the relative follows on a question 
asked by another, and virtually: gives an 
affirmative answer :— 

Mil. Gl. 62 
Itane aibat tandem? Quaen me ambae 

obsecraverint. 

Cf. quae, 7b. 973 and 984: perhaps in quaen 
me n is due to dittography. 

In the remaining cases the relative 
with -ve adds an objection or protest, with 
reference to a question or proposal of the 
speaker. 

Rud. 861, 3, 6, 23 

A. Quid ego deliqui? B. Rogas? 
Quin arrabonem a me accepisti, ce. 

Bacch. 2, 3, 98 ; 

A. Sed istic Theotimus divesnest ? B. Etiam 

rogas 4 
Quin auro habeat soccis suppactum solum. 
(se. rogasne de illo qui &c.: qui MSS.: quin 

Bergk, Ritschl.) 

Similarly Cat. 64, 180, 7. 183; Aen. 
4, 538 quiane, 10, 673 quosne. 

When quine &c. follow on a question 
asked by the speaker (Cat. 64, 180), it is 
not always easy to decide whether the rela- 
tive clause really forms a distinct question 
any more than does the third line of Dido’s 
speech (Verg. Aen. 4, 534) 

En, quid ago? rursusne procos inrisa 
priores 

Experiar, Nomadumque petam conubia sup- 
plex 

Quos ego sim totiens iam dedignata mari- 
tos ? 

But we can hardly suppose that the force 
of -ne was different from that which it has 
in e.g. Cic. De Or. 3, 214 (quoted by Ellis 
on Cat. 64, 178), ‘quo me miser conferam ! 

NO, LXXY, VOL. 1X, 


17 


in Capitoliumne? At fratris sanguine re- 
dundat. An domum ? matremne ut wiseram 
lamentantemque videam? It is significant 
that a relative with -ne is never found 
simply carrying on and confirming an 
assertion of another speaker. 

Was this particula confirmativa origin- 
ally the only enclitic ne, and is the negative 
sense of the particle in quin merely the 
result of a confusion with the proclitic ne 
of noenu, nevis, nevolt, negotium, &ec., a 
confusion that ultimately caused -ne to be 
confined to questions, in which -ne is either 
equivalent to nonne, or is a mere particle of 
interrogation, not suggesting an aflirma- 
tive any more than a negative answer? 
Certainly in Plautus -ne in questions is not 
so often equivalent to nonne as one might 
be led to suppose. Thus in 7rin. 129, 136, 
the question with -ne differs from the 
numerous Plautine questions without inter- 
rogative particles only by an added emphasis. 
Bacch. 91, 1, 1, 58 ‘Sumne autem nihili, 
qui nequeam ingenio moderari meo 1!’ =‘ Am 
I really so weak as to be incapable of con- 
trolling myself?’ Pistoclerus is yielding, 
and he begins to have greater confidence in 
his power of resisting temptation. Bacch. 
561, 3, 6, 32 ‘misine’=‘ did I send you’ 
The question is put in a formal, impressive 
way. ‘This emphasis explains the imperative 
force of manesne Most. 887 ‘ are you remain- 
ing?’ : an impatient question is equivalent to 
acommand. Contrast Mil. Gl. 57 ‘hicine 
Achilles est?’=‘Is this indeed Achilles?’ 
with Amph. 1, 1, 252, where we have non 
(nonne MSS.) and the stress is on the 
negative: yet -ne=nonne, Most. 362 and 
396. 

In conclusion the three Horatian instances 
in Minto Warren’s paper deserve a few 
words. Sat. ii. 3, 97 Sapiensne is well ex- 
plained as a question in Palmer’s edition : 
and -ne interrogative must also be found in 
Sat. i. 10, 21, where Palmer’s parenthesis 
(for are you not so to be named) is surely of 
the impossible kind, while the seeming 
parallels are false readings. Quine, if not 
the instrumental = how, may be the dative 
of the interrogative, v. Ellis on Cat. 1, 1, 
or cui may be read: the dative would depend 
on difficile. In Sat. i. 1, 108 nemon ut 
could only be ‘an absurd exclamation’ 
(Palmer) but no satisfactory alteration 
seems to be forthcoming. Could we adopt 
the conjecture quia mentioned by Palmer? 
If we read nemo quia with a colon at redeo, 
a comma at laboret, and a comma at obstat, 
the passage would run as follows: ‘Ireturn 
to my starting point: Since, as we said 

c 


18 THE CLASSICAL REVIEW. 


(V.B. probet subj.), every one is discontented 
with his own lot and praises his neigh- 
bour’s, accordingly every one has ‘ever a 
richer than himself before him, like the 
charioteer &c.’ From nemo qui(a) avarus 
might come nemo ut avarus ; cf. Palmer on 
Amph. 3, 3, 13 (qui restored for ut of B, D): 
then nemon ut might be a deliberate alter- 
ation, or a further error ; cf. Amph. 4, 3, 4 


quin utrifor qui utri. Contamination might 
be responsible for the readings mentioned by 
Cruquius, p. 314, ‘quartus autem Bland. et 
antiquiss. habet, gui nemo ut avarus quae 
lectio mihi videtur optima. In Divaei 
codice 76 qui ponitur supra ro ac: postea, ac 
potius laudet.’ 
C. M. Munvany. 


THE EI OF «i 3’ aye. 


Ove difficulty with regard to connecting 
this ef with the Conditional «i is the 
meaning: ‘but if you must come’ or ‘ but, 
if you (are willing), come’ are both un- 
natural ; another difficulty is the form, for, 
whereas we find ai beside «i and aife beside 
cide &c., I believe that «i 8 dye and not ai & 
aye is regularly found in the MSS. 

Looking at the development of Impera- 
tives like dépe, ddd’ if, and aye itself, we 
see the possibility of ei having once been 
itself an Imperative, but having become 
later on eg. merely a means of calling 
attention, not unlike our ‘Come!’ It need 
not seem strange that one Imperative should 
be joined to another, for the first Imperative 
may have come to be felt as a mere 
‘mechanism,’ like the Modern Greek as: 
in fact, dye can be used in connexion with 
a Plural Verb (cp. also idov) because of this 
exclamatory meaning. If, then, ei could 
become a kind of Exclamation, there would, 
on the one hand, be nothing unusual in the 
addition of aye,! and, on the other hand, we 
should have an explanation of the early 
meaning of eg. ei d€ od pev pev dkovoov eyo 
dé Ké Tow kataAdcEw ‘but come now, do you 


1 For apparent repetitions of the same element 
where really the element has become differentiated in 
one of its uses, cp. &AAd in the same sentence as 
&AdAo, and zgitwr in the same sentence as agitwr, &c.: 
on the other hand, there are many repetitions where 
one element may really reinforce and repeat the 
other—thus cp. possibly the Homeric pév Kev, as 
opposed to 5€ nev or 8 &y, &c. The two classes 
cannot be always distinguished, e.g. in a use like ef 
mep yap Te...Kar...aAAd Te Kal (//. 1, 81) itis dangerous 
to decide absolutely how far each re and xa still 
retains its early force. 


listen to me, and J for my part will tell 
you’; later on, the meaning may have 
changed slightly. 

Latin 4 nunc suggests an old form ez 
bearing to Greek i@. very much the same 
relation that es bears to Greek to-@. (with 
‘prothetic’ i): similarly we find some 
simple Imperativals in Vedic Sanskrit 
(e.g. han) which do not differ perceptibly in 
meaning from the Imperatives in -(d)hi. It 
is 1mpossible to say exactly where we can 
trace this Exclamatory «i: it would, of 
course, be tempting to use it to account for 
the xy in Protasis, e.g. ‘Come now, let him 
not do? this, (and then) he will not do 
wrong ’-> ‘if he does not do this, (then) he 
will not do wrong.’ But we must also take 
into account the Homeric form ai, which 
must be reckoned as of different origin, 
viz. the ‘Locative Feminine’ (of course 
without sexual meaning) of the Demon- 
strative Stem which appears in Sanskrit 
asya, of which Stem «i is the Locative 
‘Neuter.’ An early meaning of «i (ai) like 
‘then’ &c. may perhaps be traced, not only 
in eira ‘then,’ &., but also in some Final 
uses like ef’ ai xe riOyras, though it cannot 
safely be considered as the present meaning 
of the majority of such constructions. The 
development of the Conditional meaning is 
too lengthy a subject to be considered here. 

E. H. Mixes. 


? There is no need to call this use always Jussive 
or Optative in origin : both Moods (Subj. and Opt.) 
as well as the Fut. Ind. are capable of a neutral use 
which may be called Concessive. On the other hand 
there is no need to altogether exclude the ‘ Hortative’ 
and ‘ Wishing’ use as origins. 


SSS eee 


———— 


THE CLASSICAL REVIEW, 19 


THE MANUSCRIPTS OF PROPERTIUS., 


Ty a recent instalment of the 7'ransactions 
of the Cambridge Philological Society (vol. iv, 
Part I) Dr Postgate has issued a pamphlet 
of eighty-two pages ‘on certaia manuscripts 
of Propertius.’ It includes a collation and 
discussion of cod. Holkhamicus 333, hence- 
forth to be called L, a MS written in 1421 
and containing nearly two-thirds of the 
author, from II xxi 3 to the end ; excerpts 
and briefer notices of several other codices ; 
and a disputation on the value and relation- 
ship of Propertius’ MSS in general. My 
name is scattered through the treatise, and 
I hasten to acknowledge the invariable 
benignity with which Dr Postgate reproves 
me, sometimes for doing what I have not 
done, and sometimes for doing what it was 
my bounden duty to do. 

To make clear what follows, let me pre- 
mise that hitherto the only MSS of 
Propertius which count for anything have 
been NAFDY with f and v the correctors of 
Fand V: AF form one family (which Dr 
Postgate calls &), DV another (which he calls 
A) ; the agreement of these two families is 
signified by the letter O ;! N has something 
in common with each family but also some- 
thing derived from neither, and this third 
element appears tooinfandv. The codex 
A leaves off at II i 63, and thus through 
four-fifths of the elegies ® is represented 
only by F and N, the former of which is 
full of blunders and the latter of foreign 
elements: when the two agree they pro- 
bably give us ®’s reading, but when they 
differ we are often in doubt concerning it: 

another scion of this stock was therefore 

_ much desired. ' The new Lisa MS agreeing 

_ often with DV, often with Nfv, but still be- 
longing in the main to this family ®. 

Dr Postgate adduces (p. 37) nine places 
where he says L has a better reading or 
spelling than the other MSS. It has IIL 
xxv 7 harena,1V i 103 harenosum, vii 25 

harundine (IIL xv 33 harene should ap- 
parently be added) for the forms without 4 ; 
but since the harenosum and harundine are 
- found also in Neap. 268 which Dr Postgate 
on pp. 50 sq. maintains with reason to be a 
copy of F, these spellings are no proof of 









1 Students who use Dr Postgate’s collation should 
be warned that in noting the agreement or disagree- 
‘ment of the other MSS with L he habitually em- 
oys the letter 0 to mean FDVN, though through- 
out the rest of the pamphlet it regularly means 
)FDV as opposed to N. 


independence. The same then is to be said 
of IIL xviii 15 uicesimus for uigesimus ; the 
former is the better spelling, but there is no 
cause to doubt that the archetype had the 
latter: the same again would have to be 
said of anubin for anubim at II xi 41 if 
that were a better spelling. Three more 
examples, If xxv 17 sub limine for sub- 
limine or sub lwmine, IIL iii 52 philetea for 
philitea, xi 23 per menia for permenia, are 
explicable as emendations of the most 
obvious sort. There remains only IV vii 
65 suma eternis for summa eternis, one letter 
nearer to the true sua maternis ; and if any 
of the other readings were clearly genuine 
this might be thought genuine too: but now 
one can only point to the similar corruption 
suma tor summam appearing in this same 
MS at III ix 11. L therefore brings 
nothing new and true to the constitution of 
the text: these readings are much inferior 
in merit to the list eollected by Messrs, 
Luetjohann and Heydenreich from the now 
discarded Groninganus.” 

In the next place then, as L has nothing 
good of its own, does it confirm F, the chief 
representative of ®, in any good reading 
hitherto unconfirmed? In that portion of 
the poems which L contains, F gives about 
eighteen readings unconfirmed as yet* and 
seemingly right and genuine. Of these 
eighteen L gives only five, none important, 
Il xxii 30 num, xxvi 57 quod, Il vi 39 
consimili, LV ii 19 uoces, vi 40 umeris. It 
also supports F in one reading which may be 
right, If xxxii 13 ecreber platanis pariter ; 
and further it agrees with F alone in some 
four places where F, though not better than 
DVN, preserves or may preserve one part of 
the truth while they preserve another, IT 
xxii 50 fata, xxxii 37 non, xxxiv 53 
restabit erdpnas, LV iii 11 et pacate mihi.* 

2 Dr Postgate (p. 39) bases or essays to base a 
conjecture on an unsupported reading of L at III 
iv 22 ‘me sat erit sacra plaudere posse uia’ NV, 
media DF, uoc. om. L, where he proposes ‘me sacra 
sat erit.’ 

3 y has three of them, but that does not count as 
confirmation. 

4 There are also two places where Dr Postgate 
(pp. 39-41) builds conjectures, as he is quite entitled 
to do, on the joint testimony of F and L; but the 
conjectures themselves appear to have no advantage 
over the proposals of earlier critics, At III] xii 43 
‘Sirenum surdo remige adisse lacus’ F has latus (s 
in ras.), L /atreus, Dr Postgate accepts latus and 
alters Strenum to Sichnium: if the vulgate needs 
changing, this change is no easier than Schrader’s 
locos or than transposing lacus with the domes of the 

CZ 


20 THE CLASSICAL REVIEW. 


L therefore confirms F in very few good 
readings ; and the question arises, Is confirm 
the word to use? Is L an independent 
witness to the ® tradition, or is it merely 
derived from F? Probably the former : the 
indications are few, far fewer than Dr 
Postgate imagines, but they seem enough. 
At IV vii 92 F omits the word onus alto- 
gether, L gives, not the word onws, but the 
letter 0: the two appear to have copied from 
the same exemplar. There is moreover at 
least one place where L seems to have pre- 
served alone a corrupt reading of ®, a place, 
that is to say, where it stands midway be- 
tween the true reading of DV and the corrupt 
reading of F, and F’s reading seems to be a 
further corruption of L’s: III xix 6 /ontis 
DV (and N) rightly, fontes L, montes F. 
There is another place where L may indicate 
what corrupt reading stood in ® or even in 
O: IV x 41 Vncomanit DV, Butoniam D 
marg, Dutomani F, Vntoniant L: ® probably 
and O possibly had Vitomanz. Once perhaps 
L even preserves alone a corrupt reading of 
O: if at ILI xiii 27 we compare the dlius 
munus decussa of D (V_ has_ seemingly 
suffered correction) with the ¢//ts..... (gap of 
five letters) decussa of F, the ¢lis munus 
decussa of L looks like the origin of both. 
Apart from the above examples must be set 
the two following: II xxxiv 4 j/ormosam 
N rightly, formam L, et formam FDV ; III 
vii 68 thetis N rightly, petis L, pedis FD (V 
is erased). Here L is less corrupt than 
FDYV, but there is no reason to think that it 
preserves the reading of ® or of O: the 
consent of F with the other family is a 
reason for thinking otherwise! In these 
two instances—as perhaps in a third III xi 
14 Meotis (= Maeotis) N rightly, Iniectis 
FDV, Nectis L, where Jniectis and Nectis 
seem severally derived from Mectis—L may 
be exhibiting traces, as it does elsewhere, of 
the undiscovered source Z whence Nfv are 


hexameter. At III xvii 17 ‘dum modo purpureo 
spument mihi dolia musto’ spument is an old correc- 
tion, DVN have nwmen, F nwmine, L numeré, Dr 
Postgate offers cwmulem which is not likely to oust 
the vulgate: sé and se are both confused with n, and 
the corruption of sp is only a trifle harder ; then the 
archetype will have had nwment, altered on the one 
hand to nuwmen, on the other to nwmene whence 
numine and numeré. 

1 To show my meaning I take parallels from N: 
when at II xxxii 8 DV give rightly tibi me, N 
time, F timeo, we judge from comparing DV with F 
that N has here kept the corrupt reading of @ while 
F has corrupted it further, just as we judged at III 
xix 6 that L had kept and F corrupted the fontes of 
#; but when at IJ] vi 41 FDV give quid mihi si 
and N quod nisi et we do not infer that N preserves 
the reading of # or O, since FDV consent against it 
and N had elsewhere to draw from. 


in part derived; or after all these may be 
mere freaks of the pen, stumblings backward 
towards truth. But after subtracting these 
three passages the above evidence makes it 
likely that L independently preserves a 
small fragment of the tradition of ® and is 
not derived from F. 

I have selected most of the above ex- 
amples from a great number which Dr 
Postgate (pp. 21 sgq.) presents as proofs but 
which are not ‘proofs at all. For instance, 
he begins reasonably enough ‘If L was 
copied from F, it must have been copied 
from F corrected,’ and then, to show that it 


“was not copied from F corrected, proceeds 


‘whereas it often sides with the first hand 
against the corrector’s,’ as if that were any 
objection : MSS copied from a corrected MS 
whose first hand is not erased invariably 
waver in this way. ‘Then he goes on to 
prove that ‘Lis not derived from F in any 
stage of its existence’ by pointing out that 
L, though it often has the same omissions as 
F, does not omit de at II xxii 12, tu at xxx 
19, esé at IIL iii 24, and linque at LV ix 54, 
while F does; quite overlooking the fact 
that in three of these four places it is only 
the first hand of F which omits the words, 
while the second hand inserts them, and also 
that L’s composite text is derived in part 
from other sources, known to us, whence 
these omissions of F could easily be repaired. 
Again, Dr Postgate will have it that L can- 
not be derived from F because it often 
deserts F where F’s reading would have been 
satisfactory, so he affirms, to the scribe of L. 
That may perhaps disprove that L is copied 
straight from F; but it admits the explana 
tion, equally fatal to L, that L is copied 
from a copy of F, in which copy these 
alterations had been made by a scribe less 
easy to satisfy. The same reply invalidates 
another proof which Dr Postgate makes a 
great deal of : that Loften gives the syllable 
us instead of er, obviously from confusing 
the compendia, whereas in every place the 
syllable er is written at length in F, ‘I 
infer then,’ says Dr Postgate, ‘that Z is not 
derived from FEF’: so do I, but not from 
evidence like this. 

The use of L is therefore the following. 
Where it agrees with F, there the common 
reading of the two will generally. be the 
reading of ®; not always, for at II xxxii 8 
they both have ¢imeo, but generally. But 
where L dissents from F, that fact in itself 
tells us nothing about @ and shakes F’s 
testimony not a whit. Take examples. 
When at IT xxxii 37 DVN give hoc and FL 
non, that means that non was in ® and is 





aD IO Spi Ri agi ye ROCA 





oe Se Te 


not a mere blunder of F’s. But when at 
IV vii 25 DVNL give jfissa and F fixa, no 
conclusion can be drawn respecting ©. 
Maybe ® had /issa and F, as often, blun- 
dered; maybe ® had fia and L, as often, 
stole from the other sources: no one can 


say. Only in such a case as III xix 6 


already cited, where L stands halfway 
between F and DV, is there ground for 
thinking that it has preserved the reading of 
® against F. The adherence of L to N is 
absolutely weightless as witness to the 
tradition of ®:! both steal from other 
sources, and one of the sources whence L 
steals may very well be N itself: F remains 
the sole untainted channel of the family 
tradition when Ais absent. The adherence 
of L to DV is equally unimportant: the 
tradition of that family is seldom doubtful, 
so that a third witness is superfluous, es- 
pecially so poor a witness as L: only a case 
like IIT xxiv 33 cupias DL, cupies V, capias 
FN, where D and V are at odds, may deserve 
mention. With that exception, L should 
never be cited in an apparatus criticus unless 
it agrees with F and dissents from N, or 
else presents a reading peculiar to itself. Its 


agreements with F against N confirm F; its’ 


unique readings, though never important, 
_ are sometimes interesting and perhaps con- 
tain fragments of the lost source whence f 
and v derived what virtue they possess. 
If I were editing Propertius I should men- 
tion L in about thirty places.? 


1 Dr Postgate on the contrary says (p. 66) ‘we 
have seen that the concurrent testimony of L and N 
as to the reading of the family outweighs the dissent 
of F.’ I suppose the reference is to p. 33 where he 
quotes examples of agreement in spelling between N 
and L which he thinks must have been in @ because 
‘it is most improbable that correctors of this period 
would have troubled about trifles like egum, tinguere, 
pignera, murrea aud so forth.’ The example of f 
and vy would alone suffice to show that correctors 
write in the margin many readings which are in no 
sense corrections but merely variants, often insignifi- 
cant, sometimes senseless, which have caught their 
eye in other MSS: then the next copyist stolidly 
incorporates them in his text. But it too often 
happers that scholars, instead of acquiring by ob- 
servation a knowledge of what scribes were, prefer to 

frame from considerations of probability a notion of 
what scribes must have been. 
_* Dr Postgate occupies two pages, 35-37, in 
demonstrating what he calls the honesty of L. He 
finds that L contains few readings which can be 
imputed to the conjecture of its scribe ; he com- 
pares D, which Baehrens and J have praised for its 
honesty, and finds that D contains more readings of 
_ this sort ; and he concludes ‘in honesty then it is 
clear that L is superior to D.’ Even if the term 
_ honesty is thus restricted, the amount of such con- 
ed in D is so small that this superiority of L’s 
1s evanescent ; and an honesty which is compatible 
with such adulteration of the text as appears in L is 
t much to boast about. 


> 
7 
; 
a 












THE CLASSICAL REVIEW. 21 


The intrinsic value of L being thus 
insignificant, it occurs to Dr Postgate to 
enhance its relative value by dragging down 
F as near to the same level as may®be. I 
collected examples of F’s singular merit in 
the Journ. of Phil. xxi p. 196. Dr Postgate 
observes (p. 27) : 


In most of the instances alleged by Mr. Housman 
the difference between the various readings is very 
slight, and in many of them it seems more likely 
that F is accidentally right, has in fact blundered 
into a correction, than that one and the same corrup- 
tion should have appeared independently in all the 
other manuscripts, 


He instances III iii 11: lares F rightly, 
lacres L, lacies N, *lacres V, alacres D. In 
this particular passage—it is a pleasure to 
acknowledgeareal correction—Dr Postgate’s 
remark is just: if Jares was in ® it is hard 
to see why the intrusive and sense-destroying 
letter c appears in N and L; for those two 
MSS, though they steal from the source of 
DV, display some judgment in their thefts. 
But my list, after subtracting dares and 
the absentis falsely reported by Baehrens 
at II xxxiii 43 and some disputable ex- 
amples, numbered nineteen, and may be 
raised to twenty-one by adding II xxii 
50 quem quae F, quae quoque DVI, uers. 
om. N, and IV xi 20 windicet F, iudicet 
DVL, pag. om. N; and among these 
twenty-one there is no instance open to the 
same objection as dares. In four of them 
F is now confirmed by L, so Dr Postgate 
must confess that in those four I was right, 
and that F did not blunder into a correction ; 
though I on my part am quite willing to 
admit that this might sometimes happen. 
But I had spoken of five instances, among 
others, as very striking tokens of integrity. 
Dr Postgate objects (p. 28) : 

I find it difficult to understand why such readings, 
as ‘the recondite /wnarat’ (1V vi 25) and ‘ 7'iburtina, 
retained despite the metre,’ are such ‘very striking 
tokens of integrity.’ For some MSS they would be ; 
for F they prove nothing. F abounds in strange 
words and unmetrical lines; and, as the motive to 
alter on these grounds is completely absent, no con- 
clusion can be drawn from its not operating in a par- 
ticular case. 

My head goes round : does not Dr Postgate 
perceive that the absence of motive to alter 
strange words and unmetrical lines, and the 
abstention from such alteration which 
results from that absence of motive, are the 
‘integrity’ of F and constitute its merit? 
and that a particular case in which F has so 
abstained is a token of that integrity? and 
that a striking case is a striking token? 
Does he suppose that when I talk about 
integrity I picture the devil at the seribe’s 


22 THE CLASSICAL REVIEW. 


elbow prompting him to write lémarat for 
lunarat and the scribe prevailing against 
him by prayer and fasting? I simply mean 
that the scribe has copied faithfully what 
most scribes would have altered and the 
other scribes did alter : there stands lunarat 
saved: that is integrity. ‘And why should 
the retention of “lunarat’’ here be deemed 
more noteworthy than the corruption of 
‘“‘innabant”’ to “lunabant” in a majority of 
the MSS of Silius Italicus xii 448%’ To this 
enquiry I readily return the answer it ex- 
pects: the retention of ‘lunarat’ is not more 
noteworthy, but less. Both phenomena are 
noteworthy, but the retention by one MS of 
a rare word which its fellows have corrupted 
into a commoner one is less noteworthy than 
the corruption of a commoner word into a 
rarer. I have given the desired reply ; and 
I should like to see what Dr Postgate can 
do with it. 

But now let me point out that if Dr 
Postgate could shake the solitary witness 
borne by F he would land himself in a con- 
clusion at which he does not desire to arrive. 
In the first place, his method of degrading 
F does not really bring F’s level nearer to 
L’s: Lis degraded pari passu. If he denies 
genuineness to the unsupported readings I 
quote from F, he must in- common fairness 
renounce genuineness for the unsupported 
readings he quotes from L, which are so far 
inferior in apparent merit. In the second 
place, to depreciate F is to depreciate the 
whole family ® (as distinct from N) ; for all 
the important readings peculiar to that 
family are found in F alone. From Ii 1 to 
II i 63, where A leaves off, AF have no im- 
portant reading which is not in N or DV. 
From II i 63 to xxi 3, F stands alone and ?s 
the family, and on its sole authority rest the 
readings II x 11 carmina and xv 26 wellet (a 
‘very striking token of integrity’) which 
even the foe of the family Mr Solbisky is 
constrained to accept. From II xxi 3 on- 
wards we have F and L: L supports F in 
five or six true readings, none of them 
important, out of eighteen or so; but the 
lunarat and Tiburtina mentioned already, 
the cwmba III iii 22 which I also quoted as 
a striking token of integrity, and the eqguora 
IIT xxi 11 and aliquam IV ix 45 accepted 
by Mr. Solbisky, are not in L but in F only. 
Therefore when Dr Postgate says on p. 26 
‘ we cannot trust F unconfirmed’ he is saying 
that the family ® is practically worthless, 
though on p. 61 he commends me for main- 
taining the contrary against Mr Solbisky. 
And at the bottom of p. 28 he himself 
becomes aware whose game he is playing, 


and checks his career with the words ‘ I have 
however no wish to decry F,’ which he has 
been doing for two pages and a half.? 

Thus much I have written to adjust Dr 
Postgate’s partial estimate of his new codex 
L. But on pp. 61—74 he discusses the 
relations and comparative value of Proper- 
tius’ MSS in general. I hoped I had done 
with this matter for a long time to come ; 
for after all, Propertius’ MSS are not the 
only things in the world. But apparently, 
like Nehemiah’s builders, one must carry 
the sword to protect the labours of the 
trowel. When Baehrens Leo Solbisky and 


_I with some thought and pains have got 


this rather uninteresting garden of the 
Muses into decent order, here is Dr Postgate 
hacking at the fence for no discoverable 
reason unless it is the hope of boasting 
‘liquidis immisi fontibus apros.’ I feel it a 
hardship, but I suppose it is a duty, to 
withstand this inroad. Dr Postgate makes 
his mistakes with a tranquil air of being in 
the right which is likely enough to satisfy 
students not possessing my weary familiar- 
ity with the subject; so here I put it at 
their service. 

In confusing anew the relations of the 
MSS Dr Postgate has two principal aims: 
to exalt N and to disparage DY. It was 
easy to foresee that the next writer on Pro- 
pertius’ MSS would disparage DV : Baehrens 
had disparaged N, Mr Leo had disparaged 
O, Mr Solbisky had disparaged AF, I had 
defended one and all; so to disparage DV 
was the only way left of being original. 
Idolatry of N, on the other hand, is nothing 
new. 


1 Dr Postgate writes on p. 27 ‘Mz. Housman’s 
advocacy of the value of F’s isolated witness involves 
him in a curious inconsistency. He follows Baehrens 
in maintaining that A ‘‘is the most faithful repre- 
sentative of its family” Jowrnal of Phil. xxii p. 99. 
It certainly then ‘‘ happens curiously ” that in the 
poems in which we have both A and F, A should 
give of itself but one true reading ‘‘ solacia” I v 27, 
which Mr Housman thinks after all may be an acci- 
dent, and F three (or four), 7b. p. 100 sq.’ The 
only reason why I appear to Dr Postgate to be in- 
volved in a curious inconsistency is that he has for- 
gotten the facts, which are these. A gives a far 
greater number of truce readings than F, but 
wherever it gives a true reading that reading is also 
given either by F or by N or by both. F, which 
gives a far less number of true readings than A, 
gives two or three true readings which are given 
neither by A nor by N. I set three boys twelve 
sums: Tom does the first nine, Dick the first seven, 
and Harry the last eight ; and I say Tom has done 
most, although every sum done by him has also been 
done by Dick or Harry or both, and although the 
three last sums have been done only by Harry ; and 
I do not expect any one but Dr Postgate to tell me 
that I am thus involved in a curious inconsistency. 


| 
; 
; 
} 
i 
j 





THE CLASSICAL REVIEW. 23 


‘It is in his treatment of the Neapoli- 
tanus’ says Dr Postgate (p. 63) ‘that I 
find Mr Housman least satisfactory’; and 
he proceeds to explain why: ‘though not 
the enemy of N, he is its most discrimina- 
ting friend.’ [had said, in my discriminating 
and unsatisfactory way, that there is no 
best MS of Propertius. ‘The critics of the 
future’ writes Dr Postgate (p. 73) ‘ will, 
unless I am much mistaken, pronounce on 
the contrary that the Neapolitanus ‘s 
the best MS of Propertius, best as being 
the oldest of our witnesses ’—but age is no 
merit. Age is merely a promise of merit, 
which experience may ratify or annul. The 
hoary head is a crown of glory, says 
Solomon, 7f it be found in the way of right- 
eousness. ‘Till we have examined two rival 
MSS, we presume. that the older is the 
better. When we have examined them, we 
judge them by their contents. Till we 
have examined the Ambrosian fragment of 
Seneca’s tragedies (saec. V) and the codex 
Etruscus (saec. XI—XII) we presume that 
the former has the purer text. When we 
have examined them we find that it has 
not. Just so in the first decade of Livy: 
the MS which is by five or six centuries the 
oldest is not the best. The worst texts of 
Euripides yet known to man were written 
in classical antiquity itself. Useless then 
to call the Neapolitanus ‘ best as being the 
oldets of our witnesses,’ unless you can 
keep it out of our reach. But Dr Postgate 
continues ‘best again as the one that 
presents the greatest amount of truth with 
the smallest amount of falsehood,’ Then if 
I set a clerk to copy out the Teubner text 
the result will be in Dr _ Postgate’s 
Opinion a still better MS than the Neapoli- 
tanus, because it will present a greater 
amount of truth with a smaller amount of 
falsehood. How often must I repeat that 
the legitimate glory of a MS is not correct- 
ness but integrity, and that a MS which 
adulterates its text, as N does, forfeits 
integrity in direct proportion as it achieves 
correctness! Give us our ingredients pure : 
we will mix the salad: we will not take it 
ready made from other cooks if we can 
help it. We have the ® element pure in 
AF and the A element pure in DV and we 
can blend them for ourselves much better 
than N has blent them.!' The merits which 
I recognise in N are not the age and 
correctness which Dr Postgate expects the 
eritics of the future to admire, but these 


1 Dr Postgate disputes the proposition that N has 
borrowed from A, but I shall come to that point 
‘presently. 





two: the lesser, that it usefully supplements 
the pure but imperfect witness of AF to 
the tradition of ®; the greater, redeeming 
all its vice, that it contains in its adulter- 
ated text a third ingredient which we 
nowhere possess in a pure form, 

This brings me to speak of a cause to 
which N owes more blind worshippers than 
to either its age or its correctness, Dr 
Postgate writes (pp. 62 sg.) ‘a doubt, 
greater or less according to circumstances, 
must rest upon all unsupported lections in 
any of the manuscripts AFLDV. There 
is in fact only one known manuscript of 
Propertius whose unsupported evidence is 
to be taken into serious account in any 
considerable number of passages. I mean 
of course the Neapolitanus.’2 That is to 
say, each of the other MSS mentioned is so 
lucky as to possess a near relative which 
confirms and checks its witness: N has the 
singular misfortune to possess none. For 
this whimsical reason do many people call 
N the best MS of Propertius. Perhaps the 
simplest way to dispel the error is the 
following. Suppose that all extant MSS, 
with one exception, exhibited a text akin to 
N’s, and that the one exception were our 
D: those who now on the above grounds 
call N the best MS would then be bound 
by parity of reasoning to call D the best. 
And, I assure them, they would do so: 
they would forget all D’s faults just as they 
now forget N’s. Yet of course D would 
not really be a jot better than before. The 
confusion of thought is here: we do right 
to rejoice that we possess N rather than a 
second F or D or V; but we find a wrong 
vent for that joy when we call N the best 
MS: the proper vent is to thank providence. 
Tron is plentiful in England, so we would 
rather have the Borrowdale blacklead-mine 
than one iron-mine more; but we do not 
therefore call blacklead a better mineral 
than iron. If however any one is of opinion 
that the good readings found in N and not 
in F or D outweigh the good readings found 
in F or D and not in N, plus the excess of 
F’s or D’s integrity over N’s, then he has a 
right to call N the best MS of Propertius, 
But since ] do not see how such a compari- 
son can be carried out with any approach 
to precision I prefer to state what is roughly 
true and say that there is no best MS-. 


2 | should add F, from II i 63 onwards, because 
it is there the only respectable representative of the 
family #; but with that exception I subscribe to Dr 
Postgate. I attach little weight to F’s unsupported 
readings from I i 1 to II i 63, or to the unsupported 
readings of D which I cite J.P, xxii pp. 101—3. 


24 THE CLASSICAL REVIEW. 


Now fer the relation of N to DV. They 
have many readings in common: Baehrens 
and I account for this by the hypothesis 
that N derived them from a MS of the A 
family. This will never do if DV are to be 
brought low and N exalted, so Dr Postgate 


says (p. 66)— 


‘Now Ido not intend to examine the evidence 
which Mr Housman adduces in support of these 
statements,-—I had been pointing out how the phe- 
nomena of N’s text tallied with the hypothesis, —‘ for 
the following reason. He assumes wtthout proof that 
the common readings of N and A (DV) were derived 
by N from A, not derived by A from N nor by both 
from a common source. 
furnished, to discuss separate passages would be a 
waste of time. For what if A arose from a codex 
not differing very much from AF to start with, into 
which readings had been copied from N or some 
cognate manuscript and also from another source, 
say W, whence come the characteristic DV readings ?’ 


Very good. Now on the next page, 67, Dr 
Postgate has these words: ‘The agreements 
between N and AF (and L) are sufficient to 
warrant us in believing that N in great 
part is derived from a MS of the ® family.’ 
Suppose that some scholar, who bears to ® 
the ill will which Dr. Postgate bears to A, 
observes in Dr Postgate’s own fashion— 


‘Now I do not intend to examine the evidence 
which Dr Postgate adduces in support of his state- 
ments, for the following reason. He assumes 
without proof that the common readings of N and @ 
(AFL) were derived by N from #, not derived by & 
from N nor by both from a common source. Until 
that proof be furnished, to discuss separate passages 
would be a waste of time. For what if & arose from 
a codex not differing very much from DV to start 
with, into which readings had been copied from N 
or some cognate manuscript and also from another 
source, say K, whence come the characteristic AFL 
readings ?’ 


What reply can Dr Postgate give? None: 
he has sealed his own lips: there is not a 
pin to choose between that theory of N’s 
relation to ® and the theory he has himself 
suggested of N’s relation to A: both are 
equally possible and equally improbable. 
But observe that the only one of the two 
which occurs to Dr, Postgate is the one 
which jumps with his own prepossessions: I 
had considered and rejected both before 
ever I set pen to paper. Two obvious 
reasons against his improvised account of 
N’s relation to A are the following. First, 
the text of N, Dr Postgate himself admits 
it, is compounded from at least two 
elements, while there is no visible indication 
that the text of A contains more than one. 
What perversity then, in order to avoid 
assuming that the MS known to have blent 
two elements has blent a third as well, to 


Until that proof be | 


assume that the MS not known to contain 
more than one element contains three! 
That is the first reply: the second is this. 
The simplest hypothesis which will account 
for any given facts is held to be the likeliest 
hypothesis: it is not the practice to compli- 
cate affairs by gratuitous and unhelpful 
suppositions of ‘another source, say W,’ or 
of tive other sources, say W, U, T, 8S, R. 
At the date of its publication, what visible 
superiority had Copernicus’ account of the 
planetary movements over Ptolemy’s? Its 
simplicity : years had to pass before Galileo’s 
telescope confirmed it. The College of Car- 
dinals rejected the simple account because 
it seemed to threaten Holy Writ. Dr Post- 
gate rejects the simple account because it 
is derogatory to the scarce less sacred 
Neapolitanus. 

Again, Dr Postgate thinks, like many 
scholars, that N belongs to the 13th or 12th 
century, as, for aught I know, it may; and 
against my hypothesis, Baehrens’ rather, 
that N has borrowed from a MS of the DV 
family he writes thus (p. 65)— 


‘For the antiquity of the parent codex of DV Mr 
Housman claims only a moderate antiquity ; in vol. 
21, p. 180 note, he says ‘‘it was probably earlier 
than 1400 and certainly not much later.” Let us 
however place this codex anywhere he likes in the 
14th century ; and should N be of the 13th this 
portion of his edifice will still collapse.’ 


When I claimed for A the date 1400, that 
was merely my modesty: I claimed no 
more than I wanted for the point I was 
then discussing. But nothing ties down A 
to the 14th century or near it: it can 
easily be older than any date yet assigned 
to N: it has the valuable advantage of being 
inaccessible, so that no one can ask awkward 
questions about the date of its parchment. 
Moreover this objection invites the retort 
that, should N be older than ®, that portion 
of Dr Postgate’s edifice will collapse which 
derives N from ®. In fact, the only imag- 
inable reason why Dr Postgate does not say 
against ® everything which he here says 
against A is that he has taken no dislike 
to ®. 

Thus, when he concludes (p. 67) that 
‘the origin of the readings which N has in 
common with DV is unknown,’ any one will 
be ready with the supplement that it is 
no more unknown than the origin, which Dr 
Postgate believes to be known, of the 
readings which N has in common with AF, 
And again, when he says (p. 74) 


_ ‘the evidence of the A family must be separated 
into three : (@) evidence confirmed by @ for which O 


can be used as an algebraical expression, (b) evidence: 


Et em RR LW A yd it ling ip SA NTI tes Fit AP OMNI er Oey Malate 2 tg hte en: 


Sie ekae canes.” Cc 


< 
e 
5 





= 
7 










THE CLASSICAL REVIEW. 


confirmed by L’—pretty confirmation—‘or N, (c) 
evidence confirmed by neither. The last, though 
certainly not to be neglected, must be carefully 
sifted and received with caution until it is coufirmed 
from some undiscovered source ’"— 


there is no must about the matter except 
the necessity of gratifying Dr Postgate’s 
private enmity to A; and if he had 
conceived that enmity against © instead, he 
would here be writing 


‘the evidence of the @ family must be separated 
into three : (a) evidence confirmed by A for which 
O can be used as an alyebraical expression, ()) 
evidence confirmed by N, (¢) evidence confirmed by 
neither. The last......must be carefully sifted and 
received with caution until’ etc. 


In short, every word that Dr Postgate says 
against Baehrens’ account of the relation 
between N and A can be turned against his 
own account of the relation between N and 
®. All the tools he employs are two-edged, 
though to be sure both edges are quite 
blunt. 

But Dr Postgate further engages to show 
that DV are more interpolated (interpo- 
lated by conjecture, that is) than N. His 
method is the good old rule, the simple plan, 
of ‘heads I win, tails you lose. N and DV 
commit just the same offences : he extenuates 
them in N and denounces them in DV. 
His divers weights and divers measures may 
escape the eye in his pamphlet because they 
are there arranged on separate pages; but 
Ishall bring them together, and in juxta- 
position I faucy they will astonish even 
their owner. ; 

At III ii 1 sq., where editors read after 
Ayrmann ‘ Orphea de/enisse feras et concita 
dicunt | flumina Threicia sustinuisse lyra,’ 
FV have detenuisse, Nv detinuisse, D te 
tenuisse. I had said that the consent of F 
and V, witnesses from each family, showed 
that O had detenuisse ; and that N had made 
a bad attempt at amending the slight and 
honest error of O. Dr Postgate comments 
(p. 69) in this derisive vein :— 


* Now observe, to change one lettér and to insert a 
second and to write ‘‘ detenuisse ” for ‘‘ delenisse”’ is 
a slight and honest error. But to confound two 
letters already three times confused within the book 
of which this is the 41st line, and so alter a spelling, 
is a serious and dishonest one.’ 


These words, though meant in irony, are 
almost the literal truth. To write the ‘ uox 
nihili’ detenuisse (the term is somewhat too 
harsh but it is Dr Postgate’s own), for the 
Latin word delenisse, is a slight and a 
transparently honest error. As to the 
change of detenuisse into detinuisse, when 
Crities find a ‘uox nihili’ altered into a 


29 


Latin word they do not call it confounding 
two letters but they call it a conjecture. 
And Dr Postgate himself does as critics do 
when he has no motive for doing otherwise. 
D has here committed the very offence 
which he palliates in N: it has altered the 
‘uox nihili’ into Latin by the change of 
one letter, te fenuisse: and Dr. Postgate on 
pp. 36 sqg., where he was impugning D's 
honesty, threw this in its teeth: ‘ Now to 
turn to the honest D. It is clear that its 
scribe took an interest in his subject, and 
allowed his mind to work uponit. Hence... 
at III ii 1 the uox nihili “ detenuisse”’ 
becomes “te tenuisse.”’ There you hear 
the truth about the crime because the judge 
is no friend to the criminal. But to come 
again to p. 69: Dr Postgate goes on— 
‘Well, be it so: Mr Honsman has still to explain 
how it is that the MS which wilfully alters ‘‘ deten- 
uisse”’ here, reads ‘‘ detenere” for ‘* tenere” 


. against 
the metre at II xxx 26. 


Turn back to p. 37 and the dishonesty of 
D: ‘ Metre...is the ground for the impudent 
alteration in IV viii 58 “Teia petebat 
aquas”’ for “clamat,” the scansion Teia being 
unknown to the scribe.’ If I disputed as 
Dr Postgate disputes I should reply ‘ Dr 
Postgate has still to explain how it is that 
the MS which wilfully alters “ Teia”’ here, 
retains “Teja lucos”’ at IV viii 31’; but I 
know better. There is no discrepancy. 
Scribes are sometimes awake and sometimes 
asleep: the scribe of D was awake when he 
wrote ‘Teia petebat’ and asleep when he 
wrote ‘Teia lucos’; the scribe of N was 
awake when he wrote ‘detinuisse’ 
asleep when he wrote ‘ detenere.’ 

On p. 37 Dr Postgate quotes it as an 
instance of dishonesty in the scribe of D 
that ‘at II xxx 36 he has allowed himself 
to write “ingemuit’”’ for “accubuit” from li 
14.’ Only seventeen lines from that spot, 
at II xxx 19, the scribe of N has done just 
the same thing, has allowed himself to write 
‘inmerito’ for ‘dura paras’ from ILI xix 27.! 
Turn then to p. 68, where Dr Postgate 
enumerates ‘ the only examples in Mr 
Housman’s collection of the corruptions of 
N or outside it, in which I find the hypo- 
thesis of interpolation certain, reasonable 
or plausible’ and see if he quotes this 
passage. No; nor do I blame him, for the 
hypothesis of interpolation is neither cer- 
tain, reasonable nor plausible: I only invite 
attention to the transparent iniquity of bis 


and 


1 Bachrens, disliking N as Dr Postgate dislikes 
DV, promptly remarked ‘interpolate’: I in J./, 
xxi p. 154 resisted him as I am here resisting De 
Postgate, 


26 THE CLASSICAL REVIEW. 


procedure. On the very next page, 69, 
among places where ‘the reading of A is 
either interpolated or open to grave sus- 
picion,’ he adduces just such another, IT 
xix 26 ‘pedes’ A for ‘boues’ from I xx 8 
or IV xi 16. And much else does he 
adduce in lieu of evidence, which is not 
evidence at all but mere uncharitable im- 
putation. Here is a specimen: ‘I xix 10 
“Thessalis antiquam uenerat umbradomum” | 
A, not understanding the ace., makes the 
ghost knock, ‘“ uerberat,” before he enters !’ 
The note of exclamation is Dr Postgate’s, 
but I echo it with all my heart. ‘He even 
quotes against A’s good faith such custom- 
ary errors as ‘laudabat’ for ‘ludebat’ and 
‘externas’ for ‘hesternas.’ I declare, Dr 
Postgate’s entire observations on DV 
remind me of nothing so much as the 
famous soliloquies, described by Coleridge 
as ‘the motive-hunting of a motiveless 
malignity,’ in which Iago tries to explain to 
himself why he hates Othello. By p. 70 he 
has so incensed himself against the odious 
MSS that he finally writes— 


‘Should any one press this evidence in favour of 
the theory thrown out above that A descends from a 
corrected copy of &, I confess I do not see how he is 
to be refuted ; and when to the interpolations of A 
are added the interpolations of D, of which a portion 
have been already cited, I seem to discern some 
justification for what Mr Housman calls the grave 
and disastrous error of Lachmann in neglecting the 
Daventriensis ‘‘whose honest and independeut 
witness he mistook for interpolation.” ’ 


The tu quoque to which Dr Postgate habitu- 
ally exposes himself is once more available: 
should others collect the similar evidence 
against the good faith of ®, which he re- 
frains from collecting, and press it in favour 
of the theory that ® descends from a cor- 
rected copy of A, refutation would be 
about as difficult or about as easy. 

But Dr Postgate has thus shown to his 
own satisfaction (p. 71) ‘that A is much 
more deeply interpolated than N, and that 
where A contradicts N as to the word to be 
supplied in a lacuna of ®, N is to be believed 
rather than A.’ At Il xxiv 45 sq. ‘iam 
tibi Iasonia uecta est Medea carina | et 
modo ab infido sola relicta uiro’ DV have 
ab infido, N seruato, Fl omit the words or 
word. Dr. Postgate writes— 

‘ of course one of the two must be an interpolation. 
In deciding which, we ask first which gives the 
easier construction ; and the answer is ab infido ; and 
secondly which presents the more obvious sense and 
the answer, as we see from another supplement 
** fallaci” in D, is again ab injido.’ 

Strange, that any one could pen these words 
and not foresee the inevitable retort. If 


the supplement ‘fallaci’ shows that ab infido 
presents the more obvious sense, then it 
equally shows that serwato gives the easier 
construction. If it does not show that 
seruato gives the easier construction, then 
neither does it show that ab infido presents 
the more obvious sense. Nor indeed does 
there appear to be any tangible difference 
in obviousness between the two senses or in 
ease between the two constructions ; and I 
suggest that Dr Postgate should remodel 
his words so as to run ‘ Qf course one of the 
two must be an interpolation. In deciding 
which, we ask only which is in DV.’ But 


_Lhad said ‘what must settle the question 


in favour of ab infido for any impartial 
judge are palaeographical considerations. 
It is quite clear, as Baehrens prolegg. p. X11 
pointed out, that the scribe of the parent 
codex of the one family glanced from the 
do of modo to the do of infido and so left a 
metrical gap which F honestly preserves 
and which N fills up with the conjecture 
seruato, and I added that for Mr Solbisky 
to call ab infido a random conjecture was 
irrational. But Dr Postgate, because he 
breaks the laws of reason himself, will not 
allow any one else to keep them, and rebukes 
me as follows (p. 72) :— 


‘It is quite clear however that F and its family 
often omit words without any glancing of the eye, as 
at I xxi 5, 7 xxxiv 55 IJ] i 38 iii 21 xi 21’—he 
means 58—‘1V iii 9 xi 64, 68; and Mr Housman 
would have done well to examine the apparatus 
criticus before stigmatising as “irrational” the 
statement that ‘ab infido ” is a conjecture.’ 


Three of these nine examples which I 
should have found if I had examined the 
apparatus criticus are examples where the 
word omitted is est, i.e. no word at all but 
the single letter @ or the dotted line of the 
compendium; a fourth is qg., a fifth is te. 
But grant, as I do grant freely, that words 
are sometimes omitted without assignable 
cause not only in F and its family but in 
most other MSS: rational enquirers never- 
theless prefer suppositions which explain 
phenomena to suppositions which leave 
phenomena unexplained ; and when a word 
is missing they consider—I am ashamed to 
enunciate such truisms, but what is one to 
do {—that the likeliest cause is the recurr- 
ence of similar syllables. So does Dr 
Postgate, when rational enquiry suits his 
plan: at III iv 22 ‘me sat erit sacra plaud- 
ere posse uia’ we have a precise parallel : 
NV give sacra, FD media, L omits the 
word: Dr Postgate says (p. 39)— 


It would appear...that an epithet of ‘‘uia” was 
omitted...... Was this epithet ‘‘ sacra” or ‘‘media” ? 





4 


THE CLASSICAL REVIEW. 


Oo 
~! 


a3 so 2? Mrat it i . . . ah 
I prefer “‘sacra” for two reasons. First it is a less have sustained even that recollection. The 


obvious word to choose ; and secondly it explains 
the omission better, the scribe’s eye slipping from 
“sac” to ‘‘sat.””’ 


Shall I then reply ‘ Dr Postgate would have 
done well to examine the apparatus criticus ’ 
etc.? Not I: I commend him for following 
in this instance the dictates of reason, and 
encourage him to follow them elsewhere, 
even when it serves no cherished end. 
Another interpolation of N’s is defended 
by Dr Postgate on an earlier page, 24. At 
IfI xiv 19 sg. N has ‘inter quos Helene 
nudis capere arma papillis | fertur nec 
fratres erubuisse deos,’ O has ‘ nudis armata 
capillis’ (est armata F). I had said that 
armata was the original, ta was absorbed by 
the following pa or ca, and capere then 
inserted by N to prop the metre. L has 
since been collated, and it exhibits the very 
stage of corruption which I postulated: 
‘nudis arma capillis.’ Now hear Dr Post- 


gate :— 
‘There has occurred one of the transpositions of 
words which...... abound in Latin manuscripts. And 


the readings of O are derived from ‘‘arma cape 
capillis” (for papillis is found only in N), that of 
FDV armata by the loss of ne before pi [he means 
** the loss of pe before capi] and the change of ¢ to 
¢ (F further intruding an est), that of L by the loss 
of one ca out of two [he means ‘‘ the loss of cape”’).’ 


That zeal for N could enslave the reason 
and warp the judgment we knew; but 
apparently it can even cloud the -percep- 
tions: Dr Postgate with evident sincerity 
calls this ‘a more excellent way’ and 
seriously says that I should have adopted it 
if I had only remembered the compendium 
for per. I trust my mental balance would 


1 In the same paragraph I am told, with a 
compliment to soothe my vanity, that at II xxviii 9 
I have not sufficiently regarded manuscript abbrevia- 
tions, because, the true reading being peraeque, I 
quoted N’s per acquae as an example of its superiority 
to the other MSS, of which D has (and V_ probably 
had) pareque and F paremgue. ‘But’ says Dr 
Postgate ‘the original of the readings of all our MSS 
is neither “* pereque ” nor ‘‘ pareque” but ‘ peque” 
which L presents.” It pleases Dr Postgate to say 
so, but the statement has no other ground. That 
the readings of DVF are due to peque, which their 
common parent O had misinterpreted as pareque, is 
probable ; but | do not know what Dr Postgate 
eans by saying that peque was the original of N’s 
per aequae: N, as he is aware, derives scores of 
‘readings from an older source than O, and there is 
ot a hint that this source had peque rather than 
The peque of L no more tells for that 
opinion than the peregue of v for the contrary, But 
4 hee I concede the point, what follows? that ‘N 
is here no better than D, V or F [Dr Postgate does 
ot really mean “or F'”]; for it has misdivided the 
word, while they have wrongly expanded the 
abbreviation.’ Then N is better, because truth is 
‘More obscured by wrongly expanding the abbrevia- 















slips of the pen which I correct within 
square brackets are quite of a piece with 
the whole. 

I return to p. 72 :— 

‘In the much-canvassed omissions of N I find no 
evidence of design 111 x 17, 18 were obviously 
omitted through homoivteleuton, ‘‘ caput’ ending 
16 and 18. The omission of III xi 58’—he might 
add Il xxii 50—‘ must have been a pure accident... 
...And if so, why should we assume design at II] ix 
85% Mr Housman does not; and yet of II xxxiv 
83, one out of two places where N omits the end of 
a line, the part most liable to injury, he says that 
**the seribe saw the line was nonsense and desisted 
from finishing it.” ’ 

The places where N omits the end of a 
line are not two but five: IL xxxiv 53 
nec si post Stygias aliquid restabit erwmp- 
mas, 83 nec minor his animis aut sim 
minor ore canorus, II] i 27 Idaeum Simo- 
enta Iouis cunabula parui, v 39 sub terris 
sint iura deum et tormenta gigantum, 1V 
iii 7 te modo uiderunt iteratos Bactra per 
ortus. In one, II] i 27, the reading of O 
is certainly, in another, III v 39, possibly 
interpolated ; so that N’s omissions may 
there be placed to its credit. In the three 
others the omitted words present obvious 
difficulties ; and I inferred that N omitted 
them because of those obvious difficulties. 
It never occurred to me to reason, as Dr 
Postgate does, that a MS which omits some 
things by accident is not likely to omit 
other things on purpose: when a man is 
charged with murder it is not thought much 
of a defence to say that he has frequently 
committed homicide by misadventure. But 
to proceed : when I write ‘the scribe saw 
the line was nonsense and desisted from 
finishing it,’ Dr Postgate thinks it an 
answer to remark ‘Presumably then he was 
not the same scribe who a few pages on at 
TIL v 35 writes ‘“‘cur serus uersare boues et 
flamma boon.”’ There is no support for 
any such presumption. When Dr Postgate 
on p. 37 wrote of D ‘Metre is the ground 
for the impudent alteration in 1V viii 55 
“Teia petebat aquas” for ‘“clamat,” the 
scansion Teii being unknown to the scribe,’ 
I did not answer ‘ Presumably then he was 
not the same scribe who twenty-seven lines 
above had written “ Téii lucos’’’; nor do I 
now say that presumably the Dr Postgate 
who wrote p. 37 of this treatise was not 
the same Dr Postgate who wrote p. 72. 

tion than by misdividing the word. It will be 
observed that Dr Postgate’s zeal for N has here 
succumbed to his tenderness for L. 

On p. 23 an equally baseless charge of neglecting 
abbreviations is brought against Mr Leo and supported 
only by flat contradiction. 


28 THE CLASSICAL REVIEW. 


This flamma boon reappears on p. 66. 
Baehrens and I consider that all which N 
derives from O it derives through ® or 
through A and not through a third inde- 
pendent channel. Dr Postgate is anxious, 
in the interests of N, to believe the con- 
trary ; so he cites III v 35 where DV have 
rightly ‘cur serus uersare boues et plaustra 
Bootes, FL flamma palustra, N flamma 
boon, and comments as follows :— 


‘Unless N’s unintelligible and unmetrical reading 
was ‘‘reverentially copied” from the unknown Z, 
we must suppose it was derived either from A or 
from %, that is flamma boon from plaustra bootes or 


from famma palustra!, or elseadmit that N may be, - 


what Baehrens and Mr Housman say it is not, an 
independent witness to the reading of O.’ 


This is no way to argue, ‘unless a is 
true, either 6 or ¢ must be true, but 6 is 
absurd, therefore let us hope that c is true’: 
a logician would attempt to show some 
reason against a. But that is what Dr 
Postgate cannot even attempt; for he be- 
lieves that many of N’s readings were rever- 
entially copied from Z. Nor is this the 
only flaw in the reasoning. ‘To suppose that 
Hamma toon was derived from flamma palus- 
tra is doubtless absurd ; but it is not there- 
fore absurd to suppose that it was derived 
from ®. Baehrensand Dr Postgate and I are 
all agreed that N is an independent witness to 
the reading of @; and the phenomena here 
will be perfectly explained by supposing 
that ® had flamma boones (boones is in Par. 
8233 and Vrb. 641) with plaustra in the 
margin as a correction of flamma, and that 
in one apograph plaustra was mistaken for 
palustra and substituted not for famma but 
for boones and so produced the flamma 
palustra of FL, while in another apograph 
the correction was neglected and jlamma 
boones descended with only the loss of the 
last syllable into N. The explanation, 
though I do not pledge myself to it, is 
absolutely perfect, and Dr Postgate’s in- 
genuity, which fabricated on N’s behalf the 
wonderful scheme to justify nudis capere 
arma papillis, was quite equal to devising 
it ; but apparently he will not take so much 
trouble unless he sees hope of arriving 
thereby at some welcome conclusion. 

This ends what I have to say on Dr 
Postgate’s spirited attempt (pp. 61—74) 
to re-establish chaos amongst Propertius’ 
MSS. He calls it (p. 74) ‘a toilsome though 
necessary examination of the ‘past in 
Propertian criticism,’ The attempt to find 
grounds for groundless opinions is likely to 
be toilsome ; but the necessity seems to have 
been purely subjective. 


If it were not for the humour of the 
situation I might well resent the tone of 
placid assurance in which I, who think 
before I write and blot before I print, 
am continually admonished by the author 
of this pamphlet. Hitherto I have noticed 
only those references to myself which are 
connected with the tenour of these remarks; 
but I will here cite two more, because they 
show, better perhaps than anything yet 
quoted, what a bewildering disputant Dr 
Postgate is. The question is asked, whence 
did f and v derive their genuine readings 
which often agree with N? Dr Postgate 
writes on p. 60 that they ‘seem to be de- 
rived from a source similar to N,’ and adds 
this note: 


‘My Housman says these readings were derived 
‘*from the same lost MS whence N derives them.” 
This cannot be proved or disproved and comes in 
the end to the same thing, that is identity of source.’ 


A reader who finds this minute observation 
standing in a note all by itself, and sees 
words of mine quoted within inverted 
commas, will probably suppose that I really 
did say what Dr Postgate imputes to me. 
Perhaps I should have been right if I had; 
but it so happens that I did not. I tran- 
scribe the sentences which Dr Postgate had 
under his eyes, J.P. xxu. pp. 114 sq. : 


The simplest hypothesis is that which I have 
embodied in my stemma codicum: that f and v 
derive these elements from the same lost MS whence 
N derives them...... If however any one should prefer 
to say that f and v derive their genuine readings not 
from the same exemplar as N but from another MS 
closely resembling it, I should be unable to confute 
his opinion, just as he would be unable to substan- 
tiate it. 


Not only therefore did I not say what 
Dr Postgate represents me as saying, but I 
did say, before him, what he represents as 
being a criticism of his own, that ‘this 
cannot be proved or disproved.’ His mis- 
statement is harmless, and I acquit him of 
any intention either to garble or to 
plagiarise ; but he has done both. ~ 

The second example is more injurious. Mr 
Leo, who in 1880 denied all value to AFDV, 
said that any vulgar MS of the better sort 
would serve to check the testimony of N 
by: ‘librarii errores arguere ualebit e 
melioribus uulgaris notae libris quicumque 
eligetur.’ I said ‘ We have tried and con- 
demned the only three [Groninganus, Peru- 
sinus, Hamburgensis] among the vulgar 
MSS which have been selected by any recent 
eritic. When Mr Leo specifies his selections 
it will be possible to examine their merits: 
till then it must suffice to say that I have 





THE CLASSICAL REVIEW. 29 


scrutinised the mass of critical material 

- collected by Burmann and Hertzberg without 

_ discovering a fragment of genuine tradition 
unknown to us from NAFDVfy.’ Dr Post- 

gate (pp. 74 sq.) first imputes to Mr Leo 
an opinion which he has never expressed, 
‘that $ [=the vulgar MSS] may yet have 
some revelations in store for us,’ and then 
represents my remarks as an unsuccessful 
attempt to combat that opinion :— 


‘Mr Housman, criticising this opinion,’—which I 
had never heard of,—‘declares that ‘‘he has 
scrutinised the mass of critical material collected by 
Burmann and Hertzberg without diseovering a frag- 
ment of genuine tradition unknown to us from 
NAFDVfv.” ‘To scrutinise a collection as inaccurate 
as Hertzberg’s would appear to be a waste of time; 
and a scrutiny of Burmann’s edition did not save 

Lachmann, according to Mr Housman, from ‘‘erring 
_ grievously and disastrously.” Dismissing then this 
argument, if argument it be,’ and so on. 


~e? | 
ia 


The misrepresentation is of course uninten- 
tional and only proceeds from indistinctness 
of thought ; but could an unluckier occasion 
have been chosen for this air of triumph ? 

Neither the conception nor the execution 
of the pamphlet entitles it to so long a 
criticism ; but it is the work of a scholar 
who has done much better work before, and 
to whom Propertius and I are both of us 
considerably indebted. I should add that pp. 
42-58 give interetting information about 
various MSS, and that the excerpts from 
Parisinus 8233 and Vrbinas 641, as I con- 
jectured in J.P. xxii p. 125 that they would 
be, are even valuable and seem to show 
that the former at any rate deserves col- 
lation quite as much as L. 


A. E. Housman. 


JUVENAL, SAT. VII. 165. 


Quantum vis stipulare et protinus accipe 
quid do 
Ut totiens illum pater audiat. 


Tuis passage has baffled the ingenuity of 
commentators. The teacher of rhetoric is 
represented as complaining to a friend about 

the misery of listening to the ‘crambe 
_repetita’ of his pupil in the declamation 
class. He is willing to give any sum his 
friend likes to bargain for, on condition 
that the dreary task of listening be trans- 
_ ferred from the master to the boy’s father. 
So far all is plain. But what is the exact 
meaning of ‘et protinus accipe quid do ut 
~ totiens illum pater audiat’? What reading 
¥ is to be adopted; and what punctuation ? 
_ Are we to read ‘quid do’ or ‘quod do’ ; 
and is there to be a note of interrogation 
after ‘audiat’? ‘Quid do’ is the reading 
of P, and is so quoted by Priscian: ‘quod 
do’ is the correction of the manus altera 
in P, and appears in the majority of MSS. 
Mr. Mayor in his earliest edition read 
‘quid do,’ and defended the reading in the 
ollowing note: ‘The reading “quod do” 
easier, but ‘‘ quid” seems correct. The 
rds of the demand would be: “ Quid das 
ut toties illum pater audiat?” A third 
party asks the rhetorician (i.e. stipulatur) : 
What will you give the father to hear his 
as you do, every sixth day?” et protinus 
ipe is a parenthesis. The whole verse 











then means: “Make the demand ‘ Quid 
das &c.’ in any sense you please, and I lay 
down the amount in hard cash.”’ In the 
edition of 1880 ‘quod do’ is printed, and in 
the commentary this reading is illustrated 
by ‘accipio quod dant,’ from Cic. de fin. 
ii. § 83. A note by the late Prof. 
H. A. J. Munro is quoted, proposing to 
accept ‘quid do,’ but dealing with the words 
in a new way: ‘Quid may perhaps be right: 
quantum vis stipulare et protinus accipe— 
quid ? do ut totiens ke. [accipe] quid? says 
the one to whom the offer is made: “ receive 
what?” Then the other do ut (e.g. accipe 
ut) “why I give on condition that’’’ &e. 
Weidner, in his edition of 1873, printed 
‘Quantum vis stipulare ? en protinus accipe ! 
quin do’ &c.; that is to say, he took ‘stipu- 
lare’ as an infin., and recast the rest of the 
passage, which is translated in the note: 
‘Welche Summe willst du wetten? Siehe 
ich biete dir die Summe sofort! ja ich zahle 
dirs auf der Stelle.’ In the edition of 1889 
he returns to the ordinary reading, giving 
‘quid do’ and printing a note of interro- 
gation after ‘audiat’ (as does Biicheler-Jabn 
1886). He now translates the passage : 
‘Bedinge eine beliebige Summe, ja du 
kannst sie sofort in Empfang nehmen, ich 
gebe sie (die Summe) unter der Bedingung’ 
&ec. But there seems no trace of an interro- 
gation here: and Weidner adds: Die 
Worte sind freilich iiberflussig und wahr- 


30 THE CLASSICAL REVIEW. 


scheinlich korrupt. The note in Pearson 
and Strong’s edition (1892) is ‘Quid do is 
the reading of P., stipulare and protinus 
accipe are to be taken in close combination : 
‘ask for what you likeand get it ; what do 
I give that his father may hear him?” &ec., 
The idiom of a present tense implying a 
future meaning, “what will I give?” is 
quite common in English as in Latin.’ 

It is impossible to be satisfied with these 
attempts, which seem to be a violation both 
of Latin and English: nor does ‘quid do’ 
satisfy the ear from the point of view of 
rhythm. In the uncertainty in which the 


passage is left, I will hazard one more. 


attempt, and will suggest that the reading 
‘quid do’ is an old textual error for ‘ quid- 
dam,’ which seems to be quite possible 
palaeographically. Adopting this conjecture, 
I would translate: ‘ Bargain for as large a 
sum as you like, and take something at 
once, on condition that’ &e. This seems to 


give a force to protinus, which is generally 
left without sutlicient emphasis. The words 
are colloquial in style, and the irritated 
rhetorician may be supposed to say—‘ You 
may get any reasonable sum you please out 
of me—here, take this to go on with !—on 
condition’ &e. It will be naturally said 
that we should expect ‘aliquid’ or ‘ partem,’ 
or some such word, and not ‘ quiddam’ ; 
and this is a very just criticism. But in 
the colloquial language of Plautus we find 
‘quiddam’ used of something definite, as in 
Most. v. 1, 53 est consulere igitur quiddam 
quod te cum volo; Bacch. v. 2, 56 pudet 
dicere me tibi quiddam, where Philoxenus 
is going to tell the story of his passion. 
I shall be glad if some student of Juvenal 
will help either by corroborating this con- 
jecture or by showing its unreasonable- 
ness. 


W. W. Merry. 





LUCILIUS I. 24 (MULL), 30 (LACH). 


Tus fragment is found in Nonius 427, 4 
and reads in the MSS. :— ; 
vultus item ut facies mors cite morbus 

venenum. 

This was emended in the editio princeps 
to citra, and by Scaliger in Dousa’s ed. to 
icterw, Neither of these conjectures com- 
mends itself. Citra does not seem to 
supply the required sense, to say nothing 
of usage, and icterus is too unusual a word 
to be accepted without very conclusive 
evidence. Lachmann reads acer, for which 
Plaut. Men. 5, 2,119 morbum acrem ac durum 


can. be cited as a parallel, although modern 
editors following Spengel read acutum. 

I would suggest faeter or teter, as being 
not too far removed from the MSS. reading, 
and fitting what seems to be the meaning of 
the passage. Taeler is now considered the 
correct form, but if teter was used in the 
MSS. the corruption to citer would be still 
easier, For taeter morbus cf. Catullus 76, 
25 :— 

Ipse valere opto et taetrum hune deponere 
morbum. 
SAMUEL Batt PLATNER, 


NOTE ON INTERCALATION IN THE ATTIC YEAR, 


Thuc. iv. 118 § 12 with 119 § 1 makes 
the Athenian 14th of Elaphebolion in 
424/3 p.c. =the Spartan 12th of Geraistios. 
Thue. v. 19 § 1 identifies the Attic Elaphe- 
bolion 25th in 422/1 B.c. with the Spartan 
"Aprepiciov tetdptyn pOivovros. From these 
passages whimsically prodigious inferences 
have been drawn (see Miiller’s Handb. d. K7. 
Alt.-Wiss. vol. i. p. 750/1). The matter is 
disposed of simply enough however by 


supposing (1) that at Athens in 424/3—a 
leap-year with an intercalated month of 
twenty-nine days properly—the extra day 
was attached to Scirophorion (cf. C.J.A. ii. 
263) and that 422/1 had a day intercalated 
before Elaphebolion (C.Z.A. i. 273 gives 
us 384, 355 days for 424/3 and 423/2). Thus 
the interval between the two Attic dates 
is 722 days. (2) At Sparta there is a leap- 
month either in 424/3 or 423/2, and, as there 





THE CLASSICAL REVIEW, 


is no evidence to show that rerdpry Oivovtos 
at Sparta was (as at Athens) the 27th in 
a hollow as much as in a full month, we 
may suppose Geraistios 424/3 and Arte- 
misios 422/1 to be hollow, and the last date 


SCHENKL’S 


Epicteti Dissertationes ab Arriano digestae, 


ad fidem codicis Bodleiani, recensuit 
Henricus Scwenkt. Accedunt Frag- 


menta, Enchiridion, Gnomologiorum Epic- 
teteorum reliquiae, Indices. Teubner. 


1894. 10 Mk. 


Tue importance of this edition consists 
in the fact that it is the first which is pro- 
fessedly based upon the readings of the 
Bodleian Codex (Cod. Gr. Misc. 251).! That 
this codex is the parent of all the known 
MSS. of the Dissertations was first proved 
by the late Mr. J. L. G. Mowat in 1876, when 
he printed in the Journal of Philology (vol. 
vii. p. 60 ff.) a paper entitled ‘A Lacuna in 
Arrian.’ He there states that in collating 
_ this codex, to which his attention had been 

directed by Mr. Bywater and Mr. Hatch, 
he found one of the pages to be rendered 

partly illegible by a large smear-like blot, 
- occurring in Bk. i. ch. 18, and corresponding 
to a lacuna marked in Upton’s and Schweig- 
hiiuser’s editions and in a Paris MS, used 
by the latter, but unnoticed in the. older 
editions and in other MSS. 
___ Similar reasoning had beon employed by 
_ Valentine Rose in 1871 (Hermes v. p. 360 
_ ff.) to prove that the Bodleian MS. of the 

Vita Pythagorae, which is bound up in the 
same volume with the Epictetus, was rightly 
stated in Coxe’s Catalogue to be the source 
from which the other existing MSS. had 
been derived. Rose adds that he had been 
informed by Mr. Bywater, ina letter written 
Dec. 26, 1870, ‘dass auch in der vorherge- 
hhenden wichtigen (iltesten und bisher un- 
benutzten) Texte von Arrian’s Disserta- 
_tionen eine Rasur ganz iihnlichen Charakters 
-vorkommt.’? Mr. Schenkl informs us 
(p. xxiv.) that on reading Rose’s proof of 

the importance of the Bodleian MS. of the 


1 Not 257, as Schenkl prints at the beginning of 


* 
aoe — 


: 
















_ # A quotation from Mr, Bywater’s letter follows, 
in which two erasures are mentioned, which were 
made (as it would seem) after the later MSS. had 
been copied from the Bodleian, but no mention is 
made of the earlier stain in i. 18, 


31 
to be the 26th. We thus get 722 days 
again, the Lacedaemonians making one 


year contain 383, the other 354 days, 


T. NICKLIN, 


EPICTETUS. 


Vita Pythagorae, statim mihi suspicio orta 
est haud aliter rem se habere in Arriano 
Epicteteo, et codice ipso inspecto me haud 
Falso suspicatum esse facile intellexi. Etenim 
casu quodam satis iniquo factum est ut in 
libri Bodleiani folit 25 pagina antica macula 
adspersa....haud paucae litterae quae olim 
abi legebantur, non solum obscurarentur, sed 
etiam oblitterarentur....Cum igitur in hoc loco 
omnes dissertationum LEpictetearum libri 
manuscripti lacunas exhibeant, quid certius 
quam ad unum omnes aut ex ipso Bodleiano 
aut ex etus apographis descriptos esse? 
He does not give the date of his visit to 
Oxford in this passage, but on a later page 
(Ixvii.) we read: codicem pretiosissimum 
Bodleianum a, MDCCCLXXXx1 admodum iuvenis 
non indiligenter contuli eum editione Dido- 
tiana ; deinde saepius ad eum reversus quo- 
tienscumque in Anglia commoratus sum iterum 
iterumque singulos locos examinavi. 

One would suppose from this statement 
that, till the time of Mr. Schenkl’s visit to 
Oxford in 1881, the existence of the tell-tale 
stain in the Bodleian Codex was a secret, 
not only to outsiders, but even to Oxford 
scholars and to the Librarian himself, and 
that it was left to Mr. Schenkl to discover 
it ; yet, as we have seen, it had been publicly 
announced by Mr. Mowat five years earlier ; 
and even if Mr. Schenkl was ignorant of this, 
he could hardly fail to have been informed 
of it by the Librarian when he asked leave 
to consult the MS. 

But though Mr. Schenkl makes no refer- 
ence to Mr. Mowat’s article in pp. xxiv. and 
lxvii., where he assumes to himself the 
credit of proving the fact which Rose, or 
rather Bywater, had suggested, yet he dis- 
tinctly refers to it in p. xxxii. n. in which 
he controverts Mowat’s guesses as to the 
history of the Codex, and again in p. xe. 
in which he gives Mowat’s conjectural 
filling up of the lacuna, Whether he made 
acquaintance with Mowat’s article before or 
after his visit to Oxford in 1881 is of little 
consequence. Mowat is the person to whom 
belongs the credit of establishing the 


39 | THE CLASSICAL REVIEW. 


supremacy of the Bodleian Codex of Epic- 
tetus, and Mr. Schenkl has been guilty of 
an unpardonable breach of the comity of 
scholars in endeavouring to filch this credit 
from him. 

Passing on from the personal question, 
we have in the volume before us: 


(1) The Introduction, containing a chapter 
on the Life and Times of Epictetus and the 
History of the Dissertationes (pp. 1.—xiv.), 
Zestimonia Veterum de Epicteto (xv.—xxili.), 
an account of the MSS. of the Dissertationes 
(xxiv.—l.), a comparison between the 
readings of the Bodleian and the quotations 
from Epictetus in Stobaeus and other ancient 


writings, also between the quotations found 


in Epictetus and the originals (l.—1]xii.), 
editions of Epictetus (lxii.—lxx.), scholia 
copied from the Bodleian Codex by T. W. 
Allen (Ixx.—lxxxiv.), Adnotationis Supple- 
mentum (|xxxiv.—cxxil.) containing emen- 
dations by Reiske, Coraes, Elter and others, 
as well as further notes and corrections by 
the editor. 

(2) Text with critical notes (pp. 1—400). 


(3) Appendix containing Fragmenta Dis- 
sertationum (pp. 403—423), Lnehiridion 
(424—460), Gnomologium Hpicteteum Stobaet 
(461—478), Sententiae Codicis Vatieant 
(479—480), Moocyxlwves yvopar et trobjxas 
(481—490). 

(4) Indices Nominum et Locorum (490— 
501), Verborum (501—720). A facsimile of 
the page of the codex containing the famous 
smear is given at the end. 

The description of the Bodleian MS. 
and the faithfulness of the collation will be 
tested by Mr. Lindsay, who follows me. 
I shall confine my attention here to the 
actual Text in connexion with the Critical 
Notes and Supplement, and to the Index 
Verborum, merely remarking that the 
Introduction seems to be on the whole 
carefully written. 

In p. lxviii. the editor thus states the 
principle he has followed in framing his 
text, me imprimis id egisse ne a eodicis 
memoria sine iusta causa decederem; but he 
confesses that he finds he has too frequently 
substituted for the true reading of the 
Bodleian sive correctorum sive virorum doc- 
torum commenta: (he should certainly have 
added sive mea ipsius). In consequence we 
find the Adnotationis Supplementum full of 
retractations of the readings given in the 
text. In itself this proves perhaps nothing 
more than too great precipitancy, but the 
accepted readings, or the final emendations 
of the Supplement, are sometimes of such 


a nature as to excite grave suspicion as to 
the scholarship of the editor. 

I will begin with examples of improved 
readings due to the editor, which I have 
observed in looking through Book 1.1 

C. 7, 3.  rotTo Aeyerwoay Ott od cvykabyoe 
eis epwryow...6 orovdatos, 7 ott cuyKadels 
ovK értpeAnoerat TOD pa) €ikyn...dvactpeper Oar. 
HH ToiTwv <de> pndérepoy mpoadexopmevots 
dvaykatov 6moAoyev ott x.t-A. This is the 
ordinary text, d@ being inserted with Sd. 
Ed. omits 8¢ with S, puts a comma after 
avactpéepec Oar, and charges py to 7. 

C. 11, 28. rairov éorw éri ravtwv TO aitiov 
TOU TOLELV TL HAS 7) fy Tolety...TOD emaiper Oat 
n ovoté\Necbar 7 hevyav Twa H didKeev. 
Fortasse tod pevyew Ed. 

C. 12, 30. rovovtwv aitay <ovrwv> ovdenla 
cot didotar pnyavy ; Sb inserts ovrwv, Kd. in 
Suppl. proposes to omit airar. 

C. 15, 8. pnd’ av, eyo cor A€éyw, TpoTdoKa. 
Ed. reads otv for av found in all MSS. and 
edd. 

C. 18, 9. ei oé det rapa iow ert Tots 
aXAotpios Kakois SiatifecOo, éAéer avrov 
padrrov %» pioe. A corrector (Sc) wrote od 
which was accepted by former editors ; but 
Schenkl restores <i, on the ground that éXeos 
itself is an emotion called out by the 
sufferings of others and, as such, discouraged 
by the Stoic rule, cf. iii. 18, 7 té cot Kat To 
GAXotpiw Kak@; Ui, 22 det col wy dpyny civar, 
py bOovov, py eAcov, and i. 28, 9 ti ovv 
xareraivers adty ...0vxt dé, elrep apa, édects ; 
where eizep apa Shows that mercy is not com- 
mended unconditionally. 

C. 26, 138. rapaéas 8€ Tov dvayryve- 
cKovtTa Tovs vrobeTiKO’s Kal yeAaoavTos TOD 
brofepévov aita THY dvayvarw Zeavtov, edn, 
katayeAags. Hd reads rapagas for * * pagas S, 
kpagas Se. Perhaps we should omit kad. 
Translate ‘When he had confused the 
reader of the hypothetical arguments and 
his teacher had burst out laughing, he 
turned to him and said “ you are laughing 
at yourself.” ’ 

Some good readings may also be gleaned 
from the miscellaneous collection of emen- 
dations in the Supplement, e.g. 

C. 4, 31. Contrasting the honours paid to 
Triptolemus, the inventor of tke plough, 
with those paid to philosophers who have 
brought to light moral truths, Epict. says 
TO O€ THY GAYGeav cbpovtt...ov THv TEpt TO 
Gv adda THY mpds TO ed Lhv, Tis tuav ext 
TovTw PBwpov tdpvoato; Here editors in 

1 S stands for the original reading of the Bodleian, 
Sd.c.d. for subsequent corrections, s for the later MSS., 


Schw. for Schweighiiuser, Ed. for Schenkl, D for 
Diibner’s text in the Didot series, 


OI SAO Ras A EAD Fgh ee Ae a WA 


eee 


as yo. ee 






A SAO YY i Ce 


THE CLASSICAL 


general have followed later MSS. in 
_ reading zepi rot fjv. Reiske’s emendation 
mpos To Gpv, published for the first time in 
Schenkl’s Supplement, applies better to the 
invention of Triptolemus and gives a more 


pointed antithesis to the words which 
follow. 
C. 6, 20. aicypov corw 70 dvOpur ap- 


xerOar Kai KataAr yew Srov Kai ra aXoya, 
GXAG paGdXov evOev pev <éxeiva Karadiye> 
dpxer Gar, karaXnyelv & éd’ 3 KxaréAnkéev ed’ 
neov Kai 7 dios. Reiske’s insertion seems 
necessary to distinguish between the starting 
point (dpxerax) of man and brute. 

C. 8, 13. «i de Kands 7} qv IWAarwv... de Kd pe 
Kabripevov ¢ €KTOVELV iva Kkados yevopar.. OS TOUTO 
dvayxaiov mpos diAocodiav; Insert with 
Reiske dy after dvayxaiov. 

C. 15, 2. ovx erayyederat prroropia TOV 
_ éxtds Te TepuToujoey TO GvOpwTw ei Se py, 
éfw tu tHs idias UAns dvéEetar. Read with 
Reiske dvadééerar. 

C. 24, 6. Aeyer Ort 6 Gdvaros OUK €OTL KAKOV, 
ovde yap aia xpov" Aéyer Gre € 3 do0éia Wopos 
éori pawvopevev dvOparuv. ota be mepl mévov, 
ola b& wept ydovas, ota epi mevias eipyxer. 
Here Elter has restored ddog/a (the reading 
of §) instead of Upton’s conjecture poeta, 
and altered idov7js into ddvvys, as the context 
treats only of oe eeting trials. Cf. for the 
former e. 30, 2 dry cal pudakiv Kal deopa 
kal Odvarov Kai adogiav ri edeyes ; 


a 


_. =. 


In general however I do not think that 
the present text can be regarded as an im- 
provement on Diibner’s: many of the alter- 
ations are unnecessary, some impossible, and 
on the other hand a reading which had 
neither sense nor construction has been 
occasionally left unaltered, without even 
= warning of an obelus to the unwary 


Pe ed 


reader, 
I proceed to give examples of each 
sort. Unless otherwise stated the text and 
small-type notes are those of the present 
edition. 

C. 1, 3. dAN ore pe dv te ypaeys To 
€raipw, bet TovTwv dv ypamréov, y 
Yeap pareriy | épet: moTepov be Ypamréov TO ETALpw 
ov yparréov 7) ypappatiKi) ovK épel. 












ébrav Sh in marg. | ay S (i.e. avtTuypadys) | 
érépw: S, corr. Sd (item v. 5) | dv C. Schenkl, 
tav 8, sed fortasse gravius mendum latet | brav 
mev dvriypdpew 5én, TobTw ti ed. Trine. | ypan- 
7*’*y S (-ov 8), ypawtéwy Sd. 

Here dre has no construction and de 
TovTwy av yparréov does not give the 
required sense. In the Supplement 
Elter proposes 6 ti pév, dv Te ypddys TO 
ér., det (8c. ypadew) tov’rwv Tov ypap- 

NO. LXXV. VOL. IX. 


REVIEW 33 


which is, I 


parwr, think, right if we 
change 6 tre into dre (the true reading 
of 8, as we learn from Mr. Lindsay’s 
remarks be slow) and omit the mis- 
leading explanation in brackets, The 
editor unfortunately adds mihi iam 


dudum tale quid requir t videbatur adr’ 
OTe pev av (tamquam 6 oTav pev) Te ypadys 
TO €r. Oud Tivwv yeapparov ypamréoy. 


b 

Cut %: THY Xphow Ty éphiv tais dav 
taciats. Here the note is: fort. rav pav- 
taciav, though Upton had cited ypiors 


Téexvois and xpos yjpa from Stobaeus, and 
Schw. had referred to similar constructions 
in Polybius, 

C. 1, 13. ri oty ; 
27) yevorro. 
d€ tots Geois. 


pay TL pikpa wot datverat 


a te re 
TavTa ; apxy otv abrors ; evyouat * 


dpi 8, apxiv S, dpe? Sh, apov s | ebxov ed. 
Trine. | 38) ye UnUs s, 3h C. Schenkl. Supp). 
malin aut eB Xouau 44 Tous Beods, aut cum Schw, 
ebxaplore: (f. bx. obv vel ebyapiaTe) Tots Bevis 
seribere. 


The reading of the text seems right 
enough if ye be substituted for dé: ‘ Are 
you content with your lot?’ ‘At any 
rate I pray the gods to make me so.’ 
Cf. c. 11, 38 dxd ris ojpepov toi 
npepas ovdey ado éxitxornoroper...ovre 
Tov aypov ovTe Ta dvdpdroba...dAda Ta 
ddypara. Evxopa, ep, also iv. 1, 151. 
C. 1, 20. dAAa Kai ere tpdrepov tpooedOovTe 
T@ 'Exadpoditw [7d kvpiw| tod Népwvos xai 
dvaxptvovtt abtov brép Tod GvyKpovaOnvat * av 
tt Gédw’, pyoiv, “pO vod To Kupiw.’ 
mpoteAOdvt } S (mpowedOdvta unus s)| re 
scripst, tis S, del. s. | euplp] awedevbépy ed. 
Trinc. | delevi verba t@ wuple ut ex v. 18 
repetita. Suppl. tis temptare non debebam. 


But ts is not in the least wanted : the 
subject is Lateranus, as in the preceding 
sentence. Schw. considered it to be a 
marginal query, ‘who is he alluding 
to?’; but it is simpler to explain it as 
a dittographia of the preceding syllable. 
There seems no occasion for the article, 
as "Exadpoditw is anarthrous in c. 26, 
12 and elsewhere. Instead of r@ xupiw, 
which may have been a marginal note 
to remind the reader that Epaphr. had 
been the master of Epictetus, we 
should perhaps read Allo which would 
be easily lost after AITw, the confusion 
between the long and short vowel, and 
the loss or addition of the «¢ subser. 
being not unfrequent in this MS. 
Oo. 1, 32. droBavetv pe det. ei ibn, dro- 
Ovyokw Kav per’ éALyov, viv dpurr® Tis pas 


CMovans, era tore reOvifopa. 
Dv 


3 | THE CLASSICAL REVIEW. 
xoy wer’ C. Schenkl, kab mer? S, ef per’ ed. The text of S is quite right. Treating 
Trine. of the way in which the inteliect deals 
In a distinct alternative «i is evidently with the impressions of sense, Epictetus 
required. mentions selection, abstraction, com- 
C. 4, 24. paew Ti €oTe Gavaros, ti pvyy; position [of imaginary individuals (rdde 
Ti Berpuripioy, ri T vooxwovov, wa dvvyTat twa) such as the centaur or chimaera|, 


analogical reasoning in cases where 
there is a natural correspondence. Cf. 
Sext. Emp. cited by Schw. 
C. 6, 14. éxeivous  beV (sc. Tots ddoyors meu- 
tioned before) dpxet 70 éoOiew...xal TAAN’ 600 
eriteNed TOV GNOY WY €KacTOV. 


Néyew ev TH hvdaky “@ pire Kpitov k.7.X. 
locus varie emendari potest: ti vda<os, ti> 
kévetov, vel tl vocoKkopetoyv (voroKdmiov) coniect ; 
Tt Kovetoy vel TL Td Ka@vELOY S. 
The reference here seems to be to 
judicial penalties, with a special allusion 
to Socrates in the final words ; which 
forbids the reading vooos or VOCOKOpLOY. 
Why not tivos xwveov; ‘ Whose is . 
hemlock,’ ¢.e. ‘ What sort of person it 
was for whom (the draught of) hem- 


tay del. 3 | ddoywr scripsi, abtav S, 
The corruption in S (7év atrdy) is more 
easily explained from airéy than from 
Tov dddyov, not to mention that the 
lock was reserved?’ latter comes in awkwardly after éxetvoss. 
1 Se f awa > 73 , 

C. 6, 9. After speaking of the argument C. 7, 6. 76 Bovhopeve ev Xpyciet: Vopie nae 
from final causes in the case of sensible yy Siaxirrew a dpxet TovT0 dxovoa, dua Té TAs 
objects and our bodily appetites, Ep. con- pev Soxipous dpaxpas Vagal es d€X uD Tas 6 
tinues obdé tadra éudaiver Tov Texvitnv; ada ddoxipous d7odoKtpGLELS; Odx dpxet. 


TavTa pev <ovtw>* 7 O& roLatTn THs diavotas 
KATACKEUH...0vde TATA kava KIVHTAL...7pOs TO 
<pa> drodurety Tov TexviTHV ; 

ottw Stob., om. S, ovrw Meineke. Suppl. otrw 
quod apud Stobacum legitur obelo notat E. sic 
interpretans GAAG TadTa pey (sc. eupalver) n Se 
TolavTn T. 5. KaTaIKEUT (SC. OVK eupatver). 
Surely there cannot. be a doubt that 
ov7w With interrogative is the true 
reading: ‘Do not these things either 
prove a divine artist? If however 
these are not yet enough to prove 
it, go on to the processes of mind.’ 
Below, pi is inserted from Stob. and 
Sc: I prefer to omit it with S and to 
read aoXcirew with Stob. understanding 
it in the sense ‘to admit,’ (which is 
common in later philosophy, though 
not recognized in L. and 8.) ef. ae 
demus zrepi edoePeias (p. 84 Gomp.) « 
Kat améXeurov 70 Satpoviov, OoTEp Ol pev 
ovk améAeurov, eva, Oedv A€yovow <«ivat.... 
trAavocw 8 ws moAAov’s amoXElrovTes, 
Sext. Emp. Adv. Math. vii. 55 where 
amohelrev KataAnfw 18 opposed to 

; = 
GVOLPELV. 

C. 6, 10. (We not merely receive imprints 
from the objects of sense) dAAG kal éxdap- 
Bdvopev rt Kal adatpodpev Kal mpootiGepev Kat 
ovvtiGewev Tade Ta du’ aiTav Kat vy Ata pera- 
Baivopev am’ addrwv ex’ drdrd\a TLV a <odx> 

9 / 
OUTW TWS TAapaKEipeva. 

Tade <nal tade> Meineke, madAw Wachs- 
muth. | tia odx seripsi, 7a S, om. Stob., de 
GA’ arra cogitaverat Wachsmuth, &AAa rovTo.s 
mas Schw., GARG ovdauas Upton. Suppl. 
ovvTlOenev < ‘ais pev 80 &rAdAwy> 7a BE [ria] a0 


abtav Reiske, ovvt. rodde twa 80 adtav 
Coraes. 


0 Rey (ts 
Musonius for failing to detect an omission 
in a syllogism, replied od yap otpat, 
bnpt, To Kamiroduov kate Kava a. 


OU € 
mapadexn et amodoniudacers S, quibus correcturis 
probatis 8.dtT1 s. 


The reading of the later MSS. (dd71 
with the imperatives) makes very good 
sense: ‘Is it enough for one ‘who 
would guard against taking bad money 
for good, merely to be told that “thou 
must accept the genuine coin and 
reject the spurious’? It is not enough.’ 
Schw. has pointed out that ddr con- 
stantly bears the sense of 67. in later 
Greek, and is often written da 7é in 
the MSS. He would make a similar 
correction in c. 10, 7 éya A€yw ore arpak- 
tov éate TO Ldov; py yéevoito. GANG SLa 
Tt pels ovK eopev mpaxtixoi ; where 
however the editor keeps to the old 
reading without even mentioning Schw.’s 
certain emendation. In the present pas- 
sage dua ti is equally unmeaning. 

Epictetus, being blamed by 


ov yap oluat scripsi, odx olov ev S, ovx ofovel 
Elter | naréxavoa scripsi, natecxevaoa S, Kxaré- 
cKapas | uy yap, pny, Td) K. evérpynoa unus s. 


No scholar will deny that the reading 
ascribed to wnus s is infinitely more 
vigorous and idiomatic than any of the 
alternative readings. That of S reads, 
as Schw, has remarked, like a marginal 
interpretation. Also the word éy7é- 
mpnpy. ‘to set fire to’ is more suitable 
than xataxaiw ‘to burn down,’ and it 
is repeated in the next sentence. The 





THE CLASSICAL REVIEW, 35 


difficulty is, how should this, if it be 
the true reading, get into unus s, ie. 
into a MS. derived from S? It is too 
bold and too good for a scribe’s emen- 
dation. ‘Turning to the notes in 
Schw.’s edition, we find that the 
authority is Upton’s codex (of which I 
shall have more to say hereafter) and a 
Paris MS. in ora, which might be a 
correction from some _ independent 
source. The reading of 8 (xarecxevaca) 
may be, as Ed. conjectures, a corruption 
of xaréxavoa, which may itself have 
formed part of a marginal gloss in the 
original MS. from which S was de- 
rived. 
C, 8, 16. The good of man consists in ott 
mpoaipects | davtaciav]. 
gavtaciav <delevi, 
<xpnorixh> >. Schw. 


<kKal xpjois> 9. vel 

The phrase ypyoiws dart. occurs in 
c. 20,15 oteia 8 adyabod xpiots ota dei 
gavraciov, and in it. 1, 4 4 oteia rod 
ayabod eorw ev xpnoa pavtaciav, while 
mou. mpoaipeots is identified with the 
oveia ayaGod ini. 29, 1. The two might 
therefore be very well combined here, 
as they are in c. 30, 4 7a dya6a riva 
byty édoxer ; mpoaipects ota det Kal ypyors 
davraciov, also in ii. 22, 29: ini. 22, 
103, in which case the omission of 
kal xpjow might easily come about 
by the eye passing from the termina- 
tion of zpoaipects to that of xpyows. On 
the other hand, what explanation can be 
given of the interpolation of such a 
word as davtracidy 2 

C. 9, 11. Ep. has been grieving over the 

necessity of putting spirit into his hearers ; 
he would have preferred to have had occa- 
sion to restrain their impetuosity, pnxave- 
“pRVVoV...py Twes eurintwow Tovodtor véor<ot> 
€miyvovtes THY TpOs To's Heods ovyyEveay, Kal 
Ore Seopa twa Tatra tpoonpripeba TO copa 
‘ 4 4 a“ ec , 4 > / / 
kal Thy KTIoW...os Bapn Twa droppivar GAwor 
Kal dredGeiv pos Tovs ovyyeveis. 
- <ot> add. s. Suppl. véo <t’> E 
propter 0érwaow (an <ot y>?); sed vide an 
scribendum sit sine ulla mutatione ph éunlrrwct 
(audi els amoplas vel arvxlav) rowiTo veo 
émvyvévres ..cuvyyéeveray kal, bt (quoniam)...Bly, 


&s Bdpn...amopplyar AéAwor: facilior sane essel 


verborum conformatio si supplere liceret Kai 


<padvres> Sri vel tale quid. Nune nodum 
puto solvit E duabus litteris insertis : wh < tv > 
Ties éumlatwour. 

The reading of the text makes excel- 
lent sense with the slightest possible 
departure from 8: ‘taking measures to 
restrain the impetuosity of youngsters 
so high-spirited as to desire to depart at 





once and join the kindved deity.’ The 
deplorable emendations of the Suppl. 
are due to a misunderstanding of the 
construction. The subjunctive (6éAwor) 
follows the relative, as it does in iv. 
11, 35 dob zpeoBurns...0 tus vidv Trapabo 
padynoopevoy ecce senex cui filium in 
disciplinam tradat aliquis. — 

C. 9, 16. When God gives the signal, xai 
arodvon bpas tavtns tis irnpecias tor’ ame 
Nevaoeabe mpods adrov. 

ameAevoerbe Elter, dmoddecdbe S. 


The MS. is quite right: ‘When He 
dismisses you, then and not till then, 
depart.’ 

C. 11, 2. r&s re obv ypH ro mpdyparei 
read with Upton and Schw. ri otv; zas 
xe7, @ reading not even mentioned in the 
note. 

C. 16, 19. ov gee twa elvar rov...trép 

” ‘ ‘ ‘ ‘ 

mavrov a d0vTa Tov vuvoy Tov els Tov Oedy ; 


Siaddyra S, corr. 8, an isla aorta? 


The correction is wrongly assigned 
to s; itis due to Schw. who gives as 
the readings of the inferior codices 
diadiddv7a or diaddvta: Upton’s codex 
has d&ddvra. The Ed.’s suggestion is 
needless : AIAONTA easily passes into 
AIAONTA or AIAAONTA. 

I must be content to cite only a few out 
of the many passages I had marked in 
which the editor seems to me to have gone 
astray. I will merely note here some more 
general sources of error. 

The scribe of the Bodleian codex is very 
careless in the insertion or omission of the 
article and of dv. The editor has corrected 
this in c. 4, 23 <rd> oiwo after Schw., 6, 3 
<ra> dpara after Schw., 6, 9, %) zpobvpéa 
<i> tpos tiv cvvoveiav where the second 
is inserted on the authority of Stobaeus and 
the first 7 should probably be omitted on 
the same authority, 6, 42 [ra] éyxAyjpara 
omitted with s, and elsewhere; but he has 
failed to make a necessary correction in 2, 6 
pabetv <riv> Tod evAdyou Kai dAdyou tpoAnyw 
with s, 11, 18 <rod> devyew (aitwv) rovro 
aito dtu éd0€év cor with D, 14, 10 éréyeo@at 
id <tijs> oKias Hv H yp Tove with Stob., 16, 
12 éxi rov yuvakav...€v <r > pwr tt eyxare- 
pugev drradwrepov with D, 20, 11 éri & <rot> 
TadaiTwpov iyyemoviKod...7acav davraciay 
raparpoodexopeOa with D, 22, 6 évredée ¥ 
épxy <tijs> paxns with s, 26,5 rotrovy <rdv> 
dro\oyiopdv with D. In the same way I 
think we should correct 28, 18 (the houses 
of men and nests of storks were fired. It 


matters little: else show me) ri duaddpa 
D 2 


36 THE CLASSICAL REVIEW. 


oixia avOpurrov Kat veooows TeAapyov ws OLKYCLS. 
jporov ovv éott meAapyos Kal avOpwros; Ti 
Aéyers ; Kari TO TOpa opordtatov. [wiv OTe 
pev ék Soxav Kal Kepapidwv Kat TAWOwv oiKodo- 
peirat To oiKidia, 4 8 ex paBdwv Kat Tdod. | 
quae post duodtaror in S leguntur, Upt. post 
olknots transponendo, alii aliis modis servare 
statuerunt, ego ut glossem«a cicci. 

Upton is undoubtedly right. The 
sentence is necessary, and has not the 
least the look of a gloss. What is 
required is the insertion of the article 
between 67 and pev. I had.thought of 
7a, but from Mr. Lindsay’s remarks it 
appears that the original reading of 8 
was 6 0 é« paBdwyv (referring to weAapyos) 
instead of 4 & (referring to veocotd), 
so that we should probably read oixo- 
dome? (the -rar being due to dittography 
of ra) and insert 6 before pev, ‘the man 
builds his habitation of beams &c., the 
stork of clay.’ I think also that ri 
Agyers has got displaced from before 
OJLOLOV. 

Again with regard to ay, this is corrected 
by Ed. in 15, 8 pnd? oty eyd cou A€yw, tpoc- 
doxa, Where § has pi 6 dv, and in 11, 32 Kav 
peradogn ovk av arehevoy, Where Suppl. has 
av del. Cor., puto in archetypo scriptum fuisse 
av aT 
amehevon vel aveX., but he has left it in 11, 
27 ovdev Hv TO KWHodV oE...Tpds TO adelvar TO 
TaLolov ;...dAAQ ToLwodTdv TL GV, oOlov Kal ey 
‘Popyn twa jv TO Kwodv, Where I think we 
should either change dy into jv or at any 
rate insert jv after it, ‘it must have been 
something of the kind of which we had an 
instance in Rome.’ 

"Av is omitted where we should have 
expected it, not only in 14, 13 rive yap ddXw 
Kpeitrovi...pvrakt TapedwKev yuav Exacror ; 
where we learn from the note that dv was 
inserted by Upton and Schw., but also in 
16, 20 «i yotv anddov juny, érolovy ta THs 
dnddvos Where Coraes inserted dy, and in 29, 
D1 et 0€...€deduKer drogpavow...ti eyeydver TO 
cuvnppevm; Where there is no remark, but 
we should probably insert av. 


The editor speaks contemptuousiy of what 
is known as Upton’s codex: de codice illo, 
st dis placet, ab Uptono saepissime adhibito 
diligentius disserendum esse videtur, ne cui 
Jraudi sit fucata eius auctoritas, ut fuit 
Schweighausero &e. (p. xxvii.). But Upton 
is perfectly open in his description of what 
he calls his codex : it was a copy of the ed, 
pr. (A.D. 1535) lent to him by Harris, on 
the margin of which some readings had been 
added by an unknown scholar, who stated 


that these readings were taken from an- 
other copy of the ed. pr. which Cardinalis 
Salernitanus ad codicis Vaticani exemplar, 
quantum coniectura assequi possuin, emendart 
curaverat. ITisce vero diebus, cum annus 
ageretur a Christo nato 1548, aliwm nactus 
codicem manuseriptum e bibliotheca Cardinalis 
Carpensis, quem Albertus ille pius...w Georgu 
Vallae haeredibus...emerat, cum eo itidem, 
quod videbatur emendatissimus, hune etiam 
contuli, ac summa cura et diligentia quicquid 
inerat discriminis, nullo adhibito  selectu, 
annotavi. This second codex is still pre- 
served in the Library at Modena, where it 
was identified by Mr. T. W. Allen, and is 
described by him in the Class. Rev. ii. p. 14 ; 
but the writer of the note seems to have had 
no certain knowledge as to the first codex, 
and it has not yet been identified with any 
one of the three Vatican MSS. (64, 325, 1374) 
mentioned in Schenkl’s preface. No one has 
ventured to accuse Upton of fathering 
emendations of his own upon his ‘ codex,’ 
as Ursinus has been accused of doing, but 
Schenkl suggests that emendations of Wolf 
and other editors may have found their way 
into it. If so, it must have been at a 
period subsequent to the date named by the 
former owner of the codex, since Wolf’s 
ed. did not appear till 1560. Anyhow the 
readings which follow are not given by any 
editor before Upton. I have already said 
that the reading of this ‘codex’ in ce. 7, 32 
seems to be independent of §S, and derived 
from a better MS. The following readings 
seem to me to favour the same conclusion :— ~ 

7, 18. (If the premisses remain unaltered 
such as they were when they were conceded, 
we must abide by our concessions) raca 
dvadykyn nas ert THs Tapaxwpynoews emyrevew 
Kat TO dkdovbov adbrots mpoodéyerOat...ovde 
yap yuiv ere ovde Kah’ yas cvpBatver todTo Td 
exupepomevov, ereidi) THS Tvyxwpyoews TOV — 
Anppatov aréotynpev. [Thus printed with 
lacuna by Ed.] 


















































——— ee —— eel 


Post mpocdéxeoOa suppl. Upt. ‘ec cod.’ hace ph 
pevovtwy 5& adtay dmoia mapexwphOn Kal uas 
TATA avayKn THs Tapaxwphoews aploracda Kal — 
Tov Td GvakdAovOoyv tots adTots (advTav 
Schw.) Adyots mpocdéxecOa; ego lacunam 
indicavi. 

From Schw.’s note we learn that the 
real reading of the last clause in the 
‘codex’ was 10 dvaxéAovbov adrois Adyous 
mpoodéexerGar, which Schw. emended by 
inserting tod before ro, and reading 
Tois avtov (understanding this in the 
sense of tots judv aitov). I should 
prefer to read avrots simply, con 
sidering Adyos to be a marginal ex 


THE CLASSICAL REVIEW. 37 


: planation. But does not the fact of 
: such alterations being necessary prove 
that Upton’s filling up of the lacuna is 
not the conjectural emendation of an 
ingenious scholar, but a tradition of 
the original reading, which was _ lost 
through homoioteleuton (zpocdéxerGax) ! 
25, 4. ra oa THper ek TavTos TpoToV, TaV 
dotpiwv pH ediero. Oo TurTdv Gov...tis ov 
aderéoba divarai cov Taira. 
lacunam notavit Upt. e. cod. sed quod supplevit 
: 7d aldijuov adv vix sufficit: possis de etyvwuor, 
botov, kédomoyr al. cogitare. 
Ido not see why it should be necessary 
to mention more than two specially human 
characteristics. In c. 28, 21 76 aidijpov, 76 
‘qwTov, TO cuverov are the three selected. In 
ee 4. 20,11. 2. 4, 9. 11, iii. 17. 3, iv. 1, 161, 
_ Enchir. 24. 3, 4, we have simply these two. 
«10, 10. ris eoriv 7 Tod Koopov dioiknots Kal 
 rolav Twa Xwpav ev aAVTw EXEL TO AoyLKOV 
Gov ; 
3 


aita S, a’rgG Opt. e. cod., abri s. 

19, ll. yéyove yap otrws ro Ldovr atrod 
 &vexa TavTa Toll. Kal yap 6 nALos aiTov Eevexa 
_‘WdvTa ToL Kai TO Aourov adros 6 Zevs. 

a Aowrdy Upt. e. cod., Aurnpdy S. 


“It should have been mentioned that Upton 
on the authority of his codex also inserted 
_ dere after (Gov and changed the first rove? into 
_moeiv, in both which changes he is followed by 
Schw. It certainly reads better to have the 
aw laid down first in the infinitive, and 
then that the examples should follow in the 
indicative. 

28, 19. oddevi ody diaheper avOpwros redap- 
Dd; py yevouro GAAG Tovros od Siadéper. tive 
y duadheper ; fyrer Kal eipyoes ote aGXXAYW 
diadéepet. 

_ Here there is no note, but Upton writes 
noster codex drt dAw diadéper, which is cer- 
tainly far more vigorous : d\Aw adds nothing 
to what has been said before, and does not 
‘prepare the way for the list of qualities 
which follow. 

_ It may perhaps be worth while to mention 
hat the readings in the ‘ Codex’ are some- 
imes introduced by an tows or an dAdus, 
vhich might seem to indicate that they were 
en from a MS. which had been annotated 


y a Greek scribe. 


















_* 


Ix appending to Mr. Mayor’s review an account of Prof. Schenkl’s collation 
odleian MS. of the Dissertations I have had to depart from my original intention, I 
eant to give a full list of the instances in which Prof. Schenkl’s account of the readings 


It only remains for me now to speak of 
the Index, which is a work of immense, but 
not altogether intelligent industry. It 
the advantage over Schweighiiuser’s in con- 
taining, so far as I have been able to test it, 
every word which occurs in Epictetus and 
almost every instance of the use of each word, 
but the words are given without explanation 
and without context, except occasionally to 
show the construction; and important or pe- 
culiar uses are often smothered among the 
unimportant. Thus, under the heading év, no 
notice is taken of the use of év for eds ini. 
11, 32 év ‘Pon avépyy and ii. 20, 33 dareAGeiv 
ev Badaveiw, indeed the references themselves 
are altogether wanting, though special atten- 
tion had been called to it in Schw.’s lex. So 
the unusual form dvécraxa does not appear 
under dviornu, though in Schw.’s lex. we 
read ‘Jn praet. perf. notione activa, rariori 
usu. Bwpors mavres avOpwrot dverraxact i. 4, 
30.’ The peculiar use of épyoua ini. 1, 4 
9 Aoyiy Svvapus Toco akia otoa eAyjAvbe and 
in 1. 7, 12 éAnjAvOev uty wepl tov cvva- 
yovTwv Adywv tpaypateia is in no way dis- 
tinguished from the common use. Under 
ds c. cong. (p. 635, col. 1) i. 9, 11 of... d€Awor 
is omitted. Under érav no instance of the 
omission of the verb is noted, though stress 
was laid on this in pp. xci. and xxxix. If 
the reader is puzzled with the construction 
i.2,17 cé er dpovrilev ras av Spows Is 
tots aAAows and looks for zaés av in the 
Index, he will tind wés c. conj. and then ras 
av c. opt., but no més av c. conj.; only at the 
end of the article, half a page below, comes 
‘de wos av vide av.’ No reference is made 
to the position of the indefinite tis at the 
beginning of a sentence, as in iii. 1, 14 twa 
mor dkovw IloAguwva, 5, 6 ri wore piv yap 
mowvvra oe dec Katadndbjva. The use of 
aedov, as a particle, with the indicative is 
unnoticed in such passages as ii. 18, 15 
Opedrov tis exoysyOyn, li. 22, 12 ddhedrov eyo 
ervpecoor, ii. 21, 1 dpeAov r’yyv elyov, which 
are not even referred to at all. Under atdros 
we find the startling statement that in 
i, 25, 24 ebOds yap avros kar’ éuavtod dyAd, 
tive dAwrds eit, the reading of S (éuavrds) 
is fortasse recte. 


has 


J. B. Mayor. 


BODLEIAN MS. OF EPICTETUS. 


of the 
IT had 


of the MS. required correctio 
space. 
attributing infallibility to its sta 
instances (all taken from Book i.) which will justify this caution. 


THE CLASSICAL REVIEW. 


(1) Words wrongly omitted :— 


Schenkl. 


Epist. § 1.—rowatra. 

Cap. xi. § 3.—TovTov. 

Cap. xiv. § 7.—avOpwrivov. 
Cap. xix. Tit.—rupavvovs. 
Cap. xxv. § 4.—drav 74. 


(2) Words wrongly inserted :— 


Cap. iv. § 3.— zpos. 

§ 9.—ri &. 
Cap. xvii. § 16.—obros 6. 
Cap. xxix. § 28.—6 doyos. 


(3) Confusion of 


Epist. Tit.— eb mparrew. 

Cap. i. § 31.—apecxevakevar. 
Cap. ii. § 4.— iro. 

Cap. iv. § 1.—xkakd. 

~ Cap. v. § 2.—érohvb06 i. 

§ 9.—drorérpyrtat. 


n, but I soon found that to do this would take up too much 
I must content myself therefore with cautioning the reader of this edition against 
tements about the Oxford codex, and with giving some 


been concealed by the binding.) 


MS. 
Tau TOLAUTG. 
TOUTOU y. 
tov avOpwrivwv. 
TOUS TUPAVVOVS. 
OTav Tept TO. 


Words :— 


xaipeuw. 

TApETKEVAKE LOL. 
aro. 

Kade (A rec. in ras.) 
aroA0w67. 
drorérpyxi[ev?] (The last two letters have 


Cap. vi. § 41.— oid. pnd. 
Cap. ix. § 30.—ém. S. ert. 
Cap. x. § 2.—érayyeAAopevos. erayyeAopevos. 
Cap. xii. § 6.—éripedovpévov. eryseAomevov. 
Cap. xiv. § 16.—péuper Oar. péepacban, 
Cap. xxii. § 6.—yevéo@a. yiver Oa. 

§ 10.—eiow. €oTU. 
Cap. xxvi. § 6.—@éduv. Oédw. 

(4) Confusion of Contractions :— 

Cap. ii. § 36.—éorw. €oTal. 


Cap. vi. § 24.—éorw 78n Kat mdpeotw. 
Cup i. § 3.—érépw N. 
Cap. ix. § 30.—*7’ dddorpia arodidovs (mg.) 


” » ‘ / 

éorat Hoy Kal TaperTat. 
éeraipw S. 

Ta GANOTpia dmrodidovTos (mg.) 












Traces of at least three correctors (Schenkl’s Sd, Sc, Sd) are clearly present in the 
MS. ; and there were probably more than three. Sb and Se seem to me not later than 
the beginning of the fourteenth century. But the MS. must be examined by a specialist 
in Greek palaeography before it can be decided how many correctors there were, what 
were their dates, and which corrections should be referred to each. Since Prof. Schenkl 
himself admits his doubts about referring this or that reading to Sd, Sc or Sd, I need 


not mention cases where he seems to me to be in error, but will pass on to a matter 
of much greater moment— 





Schenkl. 
. Cap. 1. § 3.—ore. 
[ § 30.—ea0?. 
Cap. vii.§ 25.—ovyxabyjoe SN. 
Cap. xii. § 26.—olc6a. 
' Cap. xii. § 30.—é€ywv...dyvon® (corr. -és). 
Cap. xvii. § 11.—pav os. 
Cap. xxv. § 14.— ob. 
Cap. xxviii. § 18.—7 @. 
Cap. xxix. § 21.—rocotrov (corr, tocov- 
Tw). 













































Catulle et ses modéles, par Grorces Laraye, 
Maitre de Conférences 4 la Faculté des 
lettres de Paris. Hachette. 1894. 


Tuts essay is an owvrage couronné by the 
Academy of Inscriptions, and is an attempt 
to answer the question, How much is 
Catullus indebted to the Alexandrian poets, 
~ how much to the older Greek lyrists? Such 
a subject is on the face full of difficulties. 
_ The Greek lyric poets, except Pindar, exist 
_ only in fragments ; an ode as nearly complete 
as Sappho’s Ode to Aphrodite, is rare and 
exceptional. Thus the very basis of a 
really adequate judgment is wanting : for 
no one can appraise a poet fairly unless he 
has at least some one complete work befure 
him. From this point of view then the 
comparison is unequal : for the Alexandrian 
poets are well preserved, and have come 
down to us not in fragments, but entire. 
The Idylls of Theocritus, the Argonautics of 
Apollonius of Rhodes, the Alexandra of 
Lycophron, the Hymns of Callimachus, the 
astronomical poem of Aratus, have come to us 
in a form more or less complete: we can see 
perfectly the aim of each poet and the amount 
of success with which he worked it out. 
We can understand why it was that the 
Alexandrian poets so greatly influenced the 
Romans. Whereas from the short frag- 
‘ments of the lyric poets we can only arrive 
at the most imperfect apprehension of their 
distinct form ; and when the question—how 
much they affected any given Roman poet 
—is raised, can give little more than an 
approximate answer. 
There is another difficulty which besets 
the question : the absolute loss of so much 
literature which may have contributed to 


THE CLASSICAL REVIEW, 39 


(5) Passages where the original reading is ignored or wrongly given :— 


MS. 
ort (corr. dre). 
€AG(av 1) (corm. €X6e). 
ovyxabnoes S. 
olaOas (corr. -6a). 
€xels...dyvoav (corr. €Xwv...dyvoeis). 
poves (corr. pov ws). 
ov (corr. ob). 
6 8 (corr. 7 8). 


4 ~ 
TocovTwr (corr, TorovTov). 


W. M. Liypsay. 


LAFAYE ON CATULLUS. 


mould a writer, Catullus or whoever it 
may be. Any one who reads M. Lafaye's 
account of the hendecasyllable before Catul- 
lus will feel how difficult his task has been 
simply from want of materials. The hen- 
decasyllable is Catullus’ most successful 
experiment : we are curious to learn who 
had used it before him, what poems of any 
considerable length had been composed in 
this metre, and what kind of subject had 
been treated in it. But the actual specimens 
of pre-Catullian hendecasyllables are few, 
isolated, and insignificant; if there was 
nothing more than we have, Catullus may 
almost claim to be the inventor of this 
metre, so perfectly new is the development 
he has given it, and so various the tones 
which it assumes in his hands. But it is 
more than probable that specimens of 
Greek hendecasyllables were familiar to him 
which have perished entirely: and that 
these were known also to Catullus’ older 
contemporary, Furius Bibaculus. 

Mr. Lafaye’s book is open then to a criti- 
cism which is, at starting, inevitable: it 
is obliged to be tentative, conjectural, and 
in not a few cases unconvincing. Of this 
the author is himself perfectly conscious : 
in discussing the two poems on the Spxrrow, 
which have many forerunners in Greek litera. 
ture, but are essentially an original imspira- 
tion of Catullus, he cites Meleager’s well- 
known elegiacs to a locust, and even 
Archias’ epigram on a cicada killed by ants, 
and then after justly contrasting the effect 
produced by Catullus’ Passer with these 
light and somewhat frivolous effusions, con- 
cludes that it is impossible to say to what 
extent Catullus was consciously imitating 
either these or any other Greek model: ‘il 


40 THE CLASSICAL REVIEW. 


n’a point sous les yeux un livre choisi entre 
beaucoup d’autres dans sa _ bibliothéque et 
ouvert & une page determinée. I] a modelé 
sur les ouvrages des maitres, non pas son 
poéme, mais son esprit.’ Elsewhere he is 
more positive and, as I think, less satis- 
factory. For my own part I should not 
venture to say that the fragments of Calli- 
machus’ zambi are sufficient to give us an 
adequate idea of their general effect, or of 
the vtemper in which they were written : nor 
would it seem possible to me to elicit from 
a comparison of these with the fragments 
of Archilochus which of the two Catullus 
imitated. And is it not trifling to attempt, 
in the tenuity and fragmentary state of our 
knowledge, to suggest four various modes 


in which Catullus may have imitated 
Archilochus (p. 12) 4 
These are the weaknesses of a work 


which has many recommendations of style 
and treatment. The questions raised by M. 
Lafaye are very generally those which recur 
most to a student of Catullus, in other 
words, those which from a literary point of 
view are the most interesting. Such are 
the following. Had C. any predecessor in 
the pure iambic poems Phaselus ille and 
Quis hoc potest uidere? -Considering the 
frequency with which the pure iambic 
senarius occurs in Greek tragedy, it is 
remarkable that so few entire poems thus 
composed are extant. Again, in what sense 
was it true that Horace, not Catullus, was 
the first to introduce to Latium the iambi 
of Archilochus? The question is thus 
answered by M. Lafaye. Horace borrowed 
from Archilochus the metrical form of his 
Epodes, and little besides. Catullus took 
from him his vehement and choleric tone, 
his subjects, and much of his diction ; little 
or nothing of his metre. The answer is 
plausible and may be true; but as no one 
entire poem of the Parian is preserved, the 
explanation must remain a conjecture for 
want of sufficient materials. It is not even 
impossible that Archilochus wrote some 
pure iambic poems (see /r. 41, 44), and that 
these were directly copied by Catullus in 
Quis hoc potest widere? while Horace, a 
thorough Caesarian and somewhat a con- 
temner of Catullus, purposely overlooked 
this Archilochian inspiration as hateful to 
the Caesars and a thing which they would 
gladly forget or ignore. 

One of the most interesting sections of 
M. Lafaye’s work is the chapter on Catullus’ 
hendecasyllables. Here, if anywhere, the 
poet is most original, and the fatiguing 
effort of tracing the source of his inspira- 


tions is little required and only rarely 
successful. But, to make up for this, M. 
Lifaye gives us a lively and very well 
written account of the growth of the 
Roman hendecasyllable, with an abstract, 
generally more or less full, of each of the 
finer Catullian specimens. He accentuates 
the two conflicting tendencies under which 
the poet wrote them; on the one side the 
literary aspiration to reproduce for his 
countrymen the free abandon of Greek life, 
its absolute indifference to politics, its happy 
absorption in the trifling business of the 


-~moment whether love or little incidents of 


life and society ; on the other the old ethos 
of Republican Rome, which centred every- 
thing in the state and thought it unworthy 
of true Romans to obtrude upon the public 
notice any exhibition of private, especially 
erotic, feeling. It is of course the Greek 
that prevailed: but the Roman was also 
deep in Catullus, and displays itself in such 
confessions as xvi. 3 


(Jui me ex uersiculis meis putastis 
Quod sunt molliculi parum pudicum 


and 13 
male me marem putatis 


and such excuses as the famous and, since 
Catullus, again and again repeated 


Nam castum esse decet pium poetam 
Tpsum, uersiculos nihil necessest. 


It is this plain-spoken Roman feeling 
which accounts for and to some extent 
justifies the coarse language which Catullus 
habitually allows himself—the pedicabo ego 
uos et irrumabo, the pathice et cinaede, Xc., 
which defy translation and almost transcend 
the licence of Plautus and the Comic writers. 
It is impossible, here, not to think of Lucilius, 
and the loss of that first Latian Satire in 
which we should no doubt have found things 
far more shocking than anything in the 
poet of Verona. But the conflict of Greek 
and Roman tendencies which Catullus knew 
how to adjust, took in Lucilius the unfor- 
tunate line of mixing Greek and Roman 
words in a hodge-podge which has proved 
fatal to the preservation of his satires: we 
cannot compare him with Catullus in point 
of nude coarseness—the only one in which 
any comparison was possible—because the 
extant Lucilian remains are too short to 
justify it. 

M. Lafaye, in the course of this section, 
contrasts Catullus’ hendecasyllabie Minister 





5 
& 





uetuli puer Falerni with Horace’s Persicos 
odi, puer, apparatus. Each is very short, of 
j seven and eight lines respectively: each 
addressed to a young slave. But in deli- 
cacy Horace’s ode has a decided advantage : 
7 Catullus does not shrink from the ‘hardi- 
esses de l’expression’ ; Postumia is ebriosa 
— aeina ebriosior; water is called uini per- 
 nicies. Teetotalers may rejoice to think 
that this, the single poem in which Catullus 
7 makes wine his subject, is by no means one 
of the most felicitous of his attempts. How 
far more graceful is the six-line hendeca- 
: syllabic to Juventius, Mellitos oculos tuos, 
Zuuenti! None, indeed, of Catullus’ poems 
exceed in tenderness the cycle of those 
addressed to young friends of his own sex. 
_ M. Lafaye well says, in contrasting Cat. ix. 
(to Veranius just come back from Spain) 
with Hor. C. i. 36 (to Plotius Numida), ii. 7 
(to Pompeius Varus), the first ‘n’est qu'un 
cri de joie et de tendresse,’ while Horace 
scatters flowers to strew the path of his 
friends, recalls the memories which endear 
them, and dwells complacently on the festal 
preparations which await their arrival. I 
suspect that two recent articles in Philo- 
i logus by the veteran Duntzer, in which the 
claims of Horace as a lyrist against Catullus 
are strongly and even violently asserted, are 
much influenced by this and similar con- 
siderations. Certainly Horace is a much 
hearer approach to the modern ethical 
‘standard, while Catullus represents the old 
Greek, more purely pagan, and sensuous 
feeling. This question—Horace or Catullus 
the greater lyrist }—-discussed by Munro in 
a well-known chapter of his Llucidations, 
is very well treated by Lafaye, pp. 133 sgq. 
He finds the difference to lie much in 
the formal character of either poet’s lyrics. 
Catullus has no idea of a set ode, with 
its learned digressions and harmonious 
balance of parts. Each hendecasyllabic 
is the expression more or less spon- 
taneous of the feeling of the moment: 
its Greek prototype is oftener an epigram 
than a lyric. ‘This point, which is only 
lightly worked out, is likely to receive new 
ention from the increasing interest in 
Greek epigram evinced by the newly- 
published edition of the Anthologia Palatina 
by Stadtmiiller, as well as by Reitzenstein’s 
vigramm und Skolion. 
























Thetis (\xiv.), which M. Lafaye justly calls 
a misnomer, is to me less interesting. To 
‘my judgment it is unnecessarily long. On the 
Main question which it raises—-How are we 
to account for the disproportionate extent 


THE CLASSICAL REVIEW. 41 


of the episode of Ariadne betrayed by 
Theseus!—a great deal of Alexandrian 
literature is cited, and it is no doubt here, 
if anywhere, that Catullus kept close to 
Alexandrian prototypes. But it is very 
disappointing to find no allusion to the 
newly-discovered fragments of Callimachus’ 
Hecale, which throw a flood of light on the 
Callimachean conception of an Epyllion, 
and which would have supplied M. Lafaye 
with a still unransacked miue of quite new 
materials. As it is, there is little in the 
four sections into which he has divided his 
discussion (Le Conte épique) which has not 
been said before: and the comparison of 
betrayed love as depicted by Euripides, 
Apollonius Rhodius and Catullus, though 
just, is needlessly minute and even some- 
what tiresome. It would have been more 
to the point to have dwelt at greater length 
on more special details of Catullus’ 
Epyllion-—e.g. the inartistic repetition of 
particular words like mens, animus, pectus ; 
the recurrence of the same rhythmic type 
in the composition of the hexameter, so far 
removed from the variety of structure in 
the hexameters of Catullus’ Greek models, 
and specially from the //ecale, a priceless 
addition to our knowledge of Callimachus 
and full of promise for future researchers 
and wiser editors than Otto Schneider ; or 
again to have compared it with the extant 
Culex and Ciris, the latter of which is pro- 
bably the best surviving specimen of Roman 
Epyllia, and exhibits some strange poiuts of 
agreement with the earlier and of course 
far simpler work of Catullus. Perhaps too 
we might complain that the chronology of 
Ixiv., I mean the date of its composition, is 
not even touched upon. 

The section on the Elegies is slighter, but 
more interesting. It touches the question 
whether Catullus translated closely in his 
Coma Berenices ; M. Lafaye’s remarks here 
should be compared with his previous criti- 
cisms (much against me) of the poet’s trans- 
lation (li.) of Sappho’s ode Paiverad joc xivos 
tcos Ocotot. I am obliged to confess that 
his reasonings on this much-debated point 
and his animated defence of Catullus’ devia- 
tions from Sappho, as well as of the in- 
tegrity, as it stands, of the seemingly 
unconnected strophe Otium, Catulle, tihi 
molestum est, have not convinced me. As 
to the Coma, there is great reason for 
suspecting that the Latin version is not very 
close to the original ; would not Callimachus 
have thought it very inferior to his doubt- 
less elaborately finished original? In the 
discussion on lxviil. the adaptation of lyric 


42 THE CLASSICAL REVIEW. 


forms to elegy is treated at some length: 
the question of Alexandrianisms is here very 
prominent and, though the absence of any 
extant Callimachean model gives to the dia- 
tribes of a long series of scholars on Ixviil. 
an inevitable fluctuation and uncertainty, 
the really salient characteristics of the 
poem—distinguishing it from every other 
existing Roman elegy—are well brought 
out. 

The conclusion emphasizes the difference 
between the first division of Catullus’ poems 
(i—lx.) and the rest. The Alexandrian 


influence is far more perceptible in this _ 


latter and much Jarger portion than in the 
more purely lyrical. Little weight however 
can be attached to Babrens’ theory that 
Catullus began with imitating Sappho and 
the Aeolian school and only later at the 
suggestion of Calvus (why not Cinna?) came 
to Alexandrian models : equally problematic 
is the counter view that Ixv., lxvi., Ixviii. 
preceded the lyrics. Few, I think, can 
doubt that the candida diva of Ixviii. is 
Lesbia ; and if so, it must have been written 
late. 
Rosinson ELLIs. 





TYRRELL AND PURSER'’S LETTERS OF CICERO, VOL. IV. 


The Correspondence of M. Tullius Cicero. 


Edited by R. Y. Tyrreuu, Litt. 
D., Regius Professor of Greek, 
Trinity College, Dublin; and L. C. 


Purser, Litt. D., Fellow and Tutor of 
Trinity College. Vol. IV. Dublin 
University Press. 1894. 


Tue great work of Professor Tyrrell and 
Dr. Purser proceeds at a sufficient pace, 
if we allow its scope and dignity. The first 
volume appeared in 1879, and was re-edited 
in 1885; the second in 1886; the third 
in 1890. The present volume brings us 
down to Letter 544 (Fam. xiii. 16), the last 
before the death of Tullia. As there are only 
about 320 letters remaining, covering a period 
of just two years and a half—letters for the 
most part of no great length or difficulty— 
and as the need for excursuses and long 
introductions diminishes with the advance 
of the work, we may perhaps fairly expect 
that the whole will be completed in one more 
volume, to appear about 1898. The editors 
themselves must be very glad to be getting 
at last in sight of land. 

I have already, on more than one occasion, 
been permitted to express the general ap- 
preciation of this work in the Classical 
Review ; I need, therefore, now only deal 
with a few special features of the present 
volume, and note some selected passages. 


In the critical department a change has’ 


been made in accordance with the sugges- 
tions of Mendelssohn in his edition of the 
Epp. ad Fam. published last year, and very 
meekly accepted by the editors (benigne 
admoniti, p. 445). This consists chiefly in 
the substitution of P, a codex in the Paris 


Library, which was very slightingly treated 
in vol. i., ed. 2. p. 82, for T, a codex in the 
Library of Tours, described in detail, 20. p. 
78. In vol..ii. p. lx. it was maintained with 
much emphasis that T was not descended 
from P, but the editors have now very 
frankly withdrawn from their position, con- 
vinced by Mendelssohn’s arguments. The 
other changes are the discarding with con- 
tempt of a Harleian MS. used formerly, and 
the citation throughout of a Palatine MS. 
(598), which had been already described, 
vol. ii. p. Ixxxili. but is now treated as a 
more important authority (p. c.). 

It is most unfortunate that the text and 
notes seem to have been finally printed off 
and dismissed before the introductory part 
was written at all. On p. lxxxv. the editors 
warmly express their obligations to O. E. 
Schmidt’s work (1893), andsay : ‘if we had 
had the advantage of his guidance when 
originally arranging the order of the letters, 
that arrangement would have been much 
improved.’ Accordingly the order and dates 
of the letters in the text are often ad- 
versely commented on, sometimes in Part I. 
(e.g. p. liv.), sometimes in Part III. of the 
Introduction—a most perplexing and irri- 
tating arrangement, which is made worse by 
not being specially pointed out anywhere. 

The same explanation seems necessary for 
the absence of notes where they are greatly 
needed. For example, on Letter 304 (Aét. 
vii. 11), any one might well be astonished to 
find no note on tota haee Campana, which 
necessitates either a forced meaning to be 
given to haec, or the theory that Campana 
(the Med. reading for Campania) was used of 
the Roman Campagna before the Augustan 





se 


ote 


ae oF 


3 patina 2 EAN oO 


LOT FE AN ne able dL AGE Pip 


a f 


period. The discussion of this will be found 

on p. xiv., but there is no reference to it in 

the notes. Again there is a cursory ex- 

planation of Caelianuwm ilud in Letter 398 

(Att. x. 12), but it is only defended ant 

treated fully on p. xl., to which there is no 

reference in the note. So also the difficulty 
about Balbus entering the senate (Letter 

396; Att. x. 11) is only discussed on p. 

Ixvii. Surely all this confusion might well 

have been avoided. A preface is often, per- 

haps generally, the last written part, but it 
ought to be written before the subjects it is 
intended to modify are put out of hand. 

I will now select a few passages in which 
the view taken by the editors seems to call 
for some discussion. 

f Ep. 333 (Att. viii. 3, 4).—They here, quite 
rightly, I think, accept the reading invite 
(for in te) cepi Capuam, with the marginal * 
of the Med., pointing out not only how 
easily wi (here, by the way, misprinted 77) 
would fall out as a dittography after in, but 
also that Boot’s objection to imvite as not 
Ciceronian ignores the invitius of an undis- 
puted passage, De Or. ii. 364. There would 
be no definite objection to the word, even 
if it were never found. 

Ep. 340 (Att. vii. 9).—Very strong 
reasons are given why the first part of this 
letter ($ 1, 2) ‘should probably be detached 
from the succeeding part, and included in the 
collection as a substantive letter, coming 

' after Att. ix. 11 A (366).’ The main 

difficulty is to see why what is ‘ probable’ 

should not have been printed accordingly. 

Ep. 364 (Att. ix. 9).—On the disputed 
passage, twum digamma videram, there is a 
curiously inconsistent note. The editors 
say, ‘Let us add a guess. Could A have 
been written originally in mistake for the 
closely resembling A, which stands for Aourov 
or “balance” in Att. xv. 17, 1?’ This 
_ Obviously gives no explanation of the 
; is of the text, since it cannot be 
i supposed that they imagine A to be the form 
of adigamma. But at the end of a long 
note, they ‘think with Malaspina, that 
Cicero wrote diaypappa, schedule’; a per- 
fectly good word which they might very well 
have accepted into the text, and omitted the 
unmeaning ‘ guess.’ 

Ep. 365 (Att. ix. 10).—On the compara- 
tive force of iwre against recte, the editors 
take the view of Boot, after Manutius, for 
which I also have contended in my note, that 
recte must convey more moral approbation 
than iuwre. In the next sentence but one, 
they refer hune primum mortalem esse to 
Caesar, as Mr, Froude did, but (in his case) 












THE CLASSICAL REVIEW. 45 


without much appreciation of the fine balance 
of the question. The remark here, however, 
that exstingut is more appropriate for Cicero 
to use of Caesar than of Pompeius, is worth 
consideration. 

Ep. 383 (Fam. viii. 16), conturbare 
rationes.—Prof, Tyrrell quotes me as ren- 
dering ‘to make our fortunes so utterly 
bankrupt,’ and thinks that the technical 
sense need not be pressed. He is quite 
right, but unfortunately for me, at any rate, 
he seems not to possess the second and 
largely revised edition of my translation, in 
which, seven years ago, I had anticipated this 
criticism, by altering into ‘to break up 
our fortunes.’ I gladly acknowledge the 
courtesy of Prof. Tyrrell’s references to my 
translation, but it is unfortunate that, both 
here and elsewhere, he does me some in- 
justice by being seemingly not aware of my 
revised edition. 

Ep. 402 (Att. x. 16).—The Med. reading 
is novum, with nedum written above ; other 
MSS. have nedum only. The natural in- 
ference is that nedum novum was the original 
text, and that one or other word dropped 
out from similarity. But the editors, after 
Wesenberg, read non modo novum, a de- 
fensible reading, I think, but not defensible 
on the ground that nedum ‘could not be so 
used by Cicero.’ This is outdoing Boot, who 
merely says that Cicero himself does not so 
use it. This raises an important point of 
criticism of the letters. I think it should 
strongly be maintained that most of Cicero's 
correspondents were just as good authorities 
on the ordinary rules of usage as himself ; 
and that while in discussing a shade of 
meaning the usage of the author is im- 
portant, in the question of the admissibility 
of a word in Latin a letter from one Roman 
gentleman is as good as if it were from 
Cicero himself. If Balbus and Oppius (Aéé. 
ix. 7 A) used it, it was presumably good 
Latin. Moreover, the parallel of Fam. vii. 
28 is entirely ignored. I venture to think 
that in my note on the passage I have fully 
shown the appropriateness as well as the 
possibility of nedwm. 

There are several other passages which I 
had marked in reading for discussion, but it 
is impossible in such a matter to be ex- 
haustive, and I must end with some minor 
criticisms. I feel it only due to apologize 
for the trivial nature of some of these, as 
upon a work of such magnitude and ability, 
but the editors will probably be willing to 
admit that such criticisms are, when just and 
courteous, of nothing but a helpfal 
nature, 


44 THE CLASSICAL REVIEW. 


The book then is excellently printed—like 
all the Dublin University Press Series—as a 
whole, but occasionally might be improved. 
In particular, when the notes on a letter 
begin on the second column, the division of 
letters should be marked, which is not the 
case at present (see, e.g. pp. 638, 102). The 
misprints are few, and show careful reading 
for press ; of the dozen or so (mostly unim- 
portant) which I have marked, no less than 
three actually occur in the list of Corrigenda 
itself ! 

Tastes must, perhaps, be allowed ‘to differ 
about such forms as ‘ Antony’ and ‘ Pom- 
pey, even in a work addressed to scholars 


only, though the terrible hybrid ‘Cn. 
Pompey,’ might frighten any one. They 


cannot be allowed to differ about ‘ the battle 
of Pharsalia,’ which occurs passim. The 
vulgarism ‘try and’ for ‘ try to,’ which was 
supposed to have been banished to ladies’ 
novels, occurs twice (pp..xli. lii.). The 
spelling ‘ecstacy’ (p. 150), which still haunts 
the same region, may uncharitably be as- 
cribed to the printer. The unscholarly 
spelling reiicio occurs in many places (pp. 
49, 79, 80, etc.), though the correct rezcio is 
given also (pp. 161,265). ofos wérvuran (p. 
143) should be of rervicba. The forms 
ii, tis, are retained, as by most editors, but 
eis survives in one place (p. 320), and is 
strongly confirmed by being a correction for 
meis. 

I can hardly help thinking that there are 
some signs of flagging, excusable in the 
editors of a fourth large volume. This is 
not without a good effect in toning down 
the too exuberant facility in proposing new 


sang froid, bellum dozovdov, ‘a 
. ? 


‘guesses’ which was noted in the preceding 
volumes, so that nearly (not quite) all the 
proposed emendations in this seem meant 
for serious consideration, and many of them 
will probably win their way to general 
acceptance. But there is a distinct falling 
off in the energy which in the preceding 
volumes provided nearly every Greek word 
in the Letters with something like an 
adequate equivalent in form as well as in 
meaning, by the use of Latin, French, or 
some technical or proverbial phrase. There 
are many such here also, as etaropaxds, * with 
war &@ 
outrance,’ aidevtixds, ‘a bona fide statement,’ 
and so on. But numbers of the Greek words 
and phrases, such as trypeoia (p. 135), 
BeBiwrar and zpoBdAynpa (p. 289), ewros (p. 
292), rdvra wept rwavtwv (p. 434), and many 
others, are either ignored altogether, or 
rendered by some very tame and banal 
equivalent. Prof. Tyrrell cannot be allowed 
to fall off from the standard he himself has 
done so much to raise. 

In conclusion, it should be added that the 
notes are enriched by some valuable com- 
ments from Dr. J. 8. Reid, probably the 
highest English authority on Ciceronian 
Latin, which are interspersed (‘sine ulla 
solennitate, as Professors say at Oxford) 
much in the manner of Munro’s comments 
in the notes to Mayor’s Juvenal. Excellent 
instances are the comments on motwm and 
ad ceteros (Att. viii. 4), meam mansuetudinem 
(Att. vii. 5), fuere infantia (Att. x. 18). In 
nearly every case I am happy to find myself 
in complete agreement with Dr. Reid. 

G. E. JEANS. 


GUDEMAN’S DIALOGUS OF TACITUS. 


P. Cornelii Vaciti Dialogus de Oratoribus. 
Edited with Prolegomena, Critical Ap- 
paratus, Exegetical and Critical Notes, 
Bibliography and Indexes, by ALFRED 
GupEMAN, University of Pennsylvania. 
Boston, U.S.A. Published by Ginn & 
Jompany. 1894. 


THe unmerited neglect with which English 
scholars have hitherto passed over this 
treatise has at length been signally re- 
dressed; for it is only within the present 
year that the valuable edition by Principal 
Peterson has been noticed in these pages 
(C.R. March 1894, p. 106), and another on 


a considerably larger scale has now to be 
reviewed. At a time when so many ripe 
scholars are content to employ their talent 
only on short school-books, it is refreshing 
to find one who is willing to give the labour 
of many years?! in bringing together his 
own research, and that of almost all others 
who have at any time busied themselves on 
this dialogue. The amount of labour ex- 
pended by him may be inferred from the 
fact that the Prolegomena extend to 138 
pages, the text with its apparatus of critical 
notes to 55, and the commentary and 


1 The Preface states that it was begun as far back 
as 1888. 


-«@ 









indexes to no less than 390 pages, and that 
seven pages are filled by a mere enumeration 
of the previous works referred to in it. 
Nearly half the Prolegomena is devoted 
to the question of authorship; and on this 
point it will be best to quote the editor’s 
own summary! of his conclusions. ‘(1) 
3 That the testimony of the MSS. is unim- 
__ peachable. (2) That the treatise cannot 
possibly have been composed after the death 
' of Titus. (3) That this date, examined in 
the light of the ascertainable facts of the 
life of Tacitus, is free from all chronological 
or internal objections, and therefore no 
obstacle to the assumption that the Dialogus 
was written by the author to whom the 
4 MSS. assign it. (4) That the Dialogus and 
the admittedly genuine writings of Tacitus 
g reveal an attitude of mind and heart in the 
judgments and criticisms passed upon men 
and measures so remarkably similay as to 
be explicable only on the supposition of 
identity of authorship. (5) That, by the 
side of palpable stylistic divergences, there 
exist equally palpable stylistic coincidences. 
(6) That these differences in no sense mili- 
tate against the genuineness of the Dialogus, 
being demonstrably the necessary result of 
certain natural and well ascertainable 
causes which combined to shape as well as 
to change or even to destroy many stylistic 
features characteristic of the earliest 
publication of the future historian.’ 

There can be no two opinions as to the 
ability and learning by which these con- 
clusions are supported, especially the minute 
study of the book itself, and of other 
writings of Tacitus brought to bear on (4) 
and (5); but the consensus of manuscripts 
would have much more force if they could 
be traced to more than one original, and on 
the question of date, while the argument 
seems conclusive against those who would 

_ place the composition of the treatise as late 
as Nerva’s time, some further discussion 
might well have been given to the view of 
Mr. Peterson and others, placing it in an 
early year of Domitian (A.D. 83 or 84). 
This he dismisses summarily as a mere 
subterfuge to escape the alleged difficulties 
in the use of ‘iuvenis admodum,’ but it 
seems reconcilable with what is said 
(Suet. Dom. 9) of the good period of Domi- 
tian; and the rhetorical description else- 
where (Agr. 3) of his whole rule as a period 
of enforced silence may be exaggerated, or 
may mean no more than a necessity of 
abstaining from historical composition. 
Another considerable subject is that of the 
! See p. lvii. 










































THE CLASSICAL REVIEW. 45 


dramatic structure and the interlocutors. 
Professor Gudeman rejects the theory that 
the preliminary discussion on the respective 
superiority of poetry and oratory is merely 
a dramatic setting or framework to the 
main subject, and finds its appropriateness 
in the recognition of both as branches of 
‘eloquentia’ in the wide sense ; the problem 
being purposely left unsolved because in 
the literature of the day poetry was so 
intermingled with oratory that ‘each was 
regarded as essential to the other. It may 
be added that this prelude would no doubt 
have seemed much less disproportionate, had 
not so much of the main subject been lost 
tous. Heis in agreement with those who 
hold that Maternus represents the opinions 
of the author, and that he is not the sophist 
of that name put to death by Domitian, but 
had died before the treatise was written ; 
also with those who hold that Secundus 
must have taken a substantial part in the 
dialogue, and that the beginning of his 
speech as well as the end of Messalla’s are 
lost in the lacuna of c. 35; but he and Mr. 
Peterson are on opposite sides as to the 
alleged second lacuna in c. 40. He meets 
the improbability that such a hiatus should 
begin and end without a broken sentence 
by instancing the great gap in Annals, Book 
v. and maintains that Secundus is speaking 
from c. 36 to ‘admovebant’ in c. 40, 7, 
showing by an elaborate array of passages 
in parallel columns that several utterances 
in those chapters conflict with those assigned 
to Maternus in ec. 11—13 and ec. 40—41. 
Probably on this question the last word has 
not yet been said, but for the present he 
appears to be in possession of the field. 

In another chapter, given to the literary 
sources of the work, it is maintained that 
the dialogue is probably not more historical 
than those of Cicero or Plato; that, besides 
the vast known debt of Cicero’s extant 
works, much may have been drawn from 
the lost J/ortensius; that little, if any 
reminiscence of Seneca is traceable; that 
several statements respecting the history of 
oratory (some of them inaccurate) are likely 
to have been taken from the Acta of 
Mucianus; and that the treatise of Chrys- 
ippus epi raidwv dywyhs was probably the 
ultimate source of the interesting remarks 
on that subject. 

Another section deals with the style and 
language. ‘The extreme fulness of its 
scale may be inferred from the fact that the 
sixteen pages of which it consists are 
hardly more than a synopsis, with references 
throughout to the notes for illustration in 


46 HE CLASSICAL REVIEW. 


detail. It is needless to say that no other 
work on the subject approaches it in ex- 
haustiveness, but probably the majority of 
readers should be recommended to be con- 
tent with the more concise sketch of the 
most salient stylistic characteristics, and of 
their relation to those of the other works of 
Tacitus, which has been already given (page 
xlvi. foll.) in the discussion of the question 
of authorship. 

The introduction is closed by a chapter 
on the manuscripts. On this subject, Dr. 
Gudeman’s main views are already known 
from his notice in this Review (vi. 316 foll.) 
of Scheuer’s treatise, where the account of 
their relation and interdependence, their 
derivation from two lost copies (X and Y) 
of a MS. discovered in the 15th century,! 
and the grounds on which he gives prefer- 
ence to those of the Y class are briefly set 
forth. Here the whole subject is fully dis- 
cussed, with analyses of the variants of the 
leading MSS. and reasons for excluding 
others from consideration. He is at issue 
on some points with Mr. Peterson, especi- 
ally as to the Harleian MS. which he rejects 
summarily as worthless, whereas the latter 
editor had considered it worthy of careful 
study. It would be impossible here to dis- 
cuss this question ; but the apparently early 
date of this MS., its interesting early 
history, and evident relation to the ‘ editio 
princeps’ will be sufficient reason to make 
most scholars glad that attention has been 
directed to it, even if its critical value 
should be rated below Mr. Peterson’s 
estimate. 

The editor’s critical recension of the text, 
evidently to a considerable extent antici- 
pated by his dissertation on the subject in 
the American Journal of Philology (xii. 327 
—347 and 444—457), may be generally 
described as conservative; the readings 
introduced on his own conjecture into the 
text being less numerous than those of 
many of his predecessors, and the MS. text 
being frequently adhered to where most 
others have altered it, as may be exempli- 
fied in one or two strong cases. (1) Ine. 
11, 9, he is almost the only editor who 
retains ‘in Neronem’ (taken with ‘ poten- 
tiam’), a construction no doubt in itself 
sufficiently Tacitean, but here harsh by 
reason of the number of intervening words. 
Is it not also doubtful whether so strong an 


1 Whether Dr. Gudeman is right in rejecting (with 
Voigt) the testimony of Pontanus, ascribing the dis- 
covery of this MS. to Henoch of Ascoli, is a point 
of small importance, and on which little can be said 
on either side. 


expression as ‘potentia in Neronem’ could 
be used of a mere freedman’s influence with 
his emperor? (2) He also defends the MS. 
text ‘si cominus fatetur’ in ce. 25, 8, which 
seems hardly possible as it stands. It may 
indeed be shown that ‘si’ could have the 
force of ‘qua,’ that ‘cominus’ could have a 
metaphorical. meaning, that ‘fatetur’ could 
mean ‘profitetur,’ but surely the ‘cominus 
agamus’ of Cicero would not justify Tacitus 
in saying ‘cominus fatetur’? Miiller’s ‘qua 
quasi cominus nisus fatetur’ may be open 
to other objections, but does at least make 
the metaphor tolerable. 

He is no doubt fully justified in saying of 
his ‘adnotatio critica’ that it ‘aims at com- 
pleteness, no emendation of any intrinsic 
value, published since Michaelis’ funda- 
mental recensio, being omitted’: whether 
he is also justified in saying of his exe- 
getical commentary that ‘no really irrele- 
vant matter has been allowed to intrude,’ 
will no doubt depend upon what his various 
readers may understand by irrelevancy. It 
should be noticed that much of its bulk is 
due to his habit of citing illustrative pass- 
ages in full, for which those who have small 
libraries or little leisure will more than 
pardon him. Allowance should also be 
made for the use of considerably larger and 
more spacious type than (for instance) that 
of the Clarendon Press, as seen in Mr. 
Peterson’s edition. This again will be 
grateful to those who have to husband their 
eyesight. For the rest, the bulk is mainly 
due to the introduction of matter as to the 
value or relevancy of which opinions will 
always differ, such as the discussion and 
refutation of views which some might think 
hardly deserving of so full a notice, or the 
collection of much lexicographical and 
grammatical matter on words and usages, 
which, however valuable in itself, should in 
the opinion of many be more fitly relegated 
to separate works, and only sparingly intro- 
duced into the commentary on a special 
treatise. On such questions all editors 
must use their own judgment, and with Mr. 
Peterson’s and Dr. Gudeman’s editions 
before them, English scholars will have no 
difficulty in choosing for themselves a com- 
mentary of the scale and character most 
suited to their requirements. I may say at 
least for those occupied, like myself, on 
other works of Tacitus, that to them hardly 
any commentary on any part of their 
author will seem too full, and that they will 
gladly welcome a storehouse of material 
tending indirectly to the furtherance of 
their own labours also. Those again who 








lee ee al daca 





are engaged ou any general study of Latin 
lexicography, grammar, textual criticism or 
the like, will recognize that there have been 
cases, and may often again be such, in which 
_ what may have been an error of judgment 
in the accumulation of an apparently dis- 
proportionate commentary on a small portion 
of the classics has led to results beyoud the 
author’s probable aim, and has borne fruit 
in assisting the permanent advance of this 
or that branch of general scholarship. The 
chief omission that few if any would regret 
and most would welcome, would be that of 
the more polemical passages, in which 
; opinions of others seem as it were paraded 
for censure and branded with notes of 
exclamation. A great mass of proposed 
emendations and interpretations are in their 
nature shortlived, and an editor who aims 
at leaving behind him a xrijpa és ded will 
often best consult the readers who most 
appreciate him by leaving the objections to 
other views to be gathered from the 
arguments for his own. 
In the selection of details for criticism 
only a few points can here be taken. One 
of the longest and principal notes is that 
on the chronological difficulty in c. 17, 10. 
Dr. Gudeman thinks that the sum of 120 
years cannot pass as a round number, and 
is not intended to be the total of the sums 
preceding it, but that what is asserted is 
that the whole period is comprised within 
_ 120 years, that being taken as the theo- 
retical maximum of human life. It seems 
doubtful whether this can be reconciled 
with the later reference in c. 24, 14, where 
the 120 years seem to be very plainly 
asserted to be the sum total of the actual 
reckoning ; and it would seem still an open 
question whether there is not less difficulty 
in the view that the total, like some of the 
details (namely, the years of Tiberius, 
Claudius, and Nero), is a round number 
slightly in excess of fact, or in the supposi- 
tion that a scribe (as in many other cases) 
has misread a figure in his exemplar, and 
written it wrongly in words.! In another 

chronological difficulty in the same chapter, 
that of the period at which Pollio and 
Corvinus were still living, he cuts the knot 
by bracketing the whole sentence as a gloss ; 
but surely the statement, that whoever had 
been a sharer in the two last ‘ congiaria ’ of 
Augustus could have heard those orators, 
‘Seems to require some such note to complete 


t 
i 
y 
, 
; 


























_ 1 Some errors now standing in the MSS., as 
-‘novem et quinquaginta,’ ‘centum et decem,’ seem 
to show that this kind of confusion has been at work 
in the passage. 


THE CLASSICAL REVIEW. 17 


it, and, granting that no eorrection can 
make the words right, Dr. Gudeman might 
perhaps see in the statement another of the 
errors for which he thinks (see on c, 34, 33) 
that the author may have been indebted to 
Mucianus. Of the readings introduced by 
himself into the text, the following appear 
to be among the most successful :-— 

(1) In ec. 1, 15, ‘excepi’ (with ‘e’ pre 
ceding) gives thoroughly the reyuired sense 
of catching up and storing in memory the 
words of a speaker, which cannot be satis 
factorily extracted from ‘accipere ab aliquo.’ 

(2) In 7, 13, ‘iuvenes vacuos’ commends 
itself as suggested by a MS. reading and 
completing the antithesis. 

(3) In 32, 28, the insertion of ‘ex’ is 
justified by its use in the passage quoted, 
and gives better sense than a simple abla- 
tive. 

(4) In 33, 18, ‘varias ac (for ‘ aut’) 
reconditas’ is simpler than any other change 
by which a right meaning is given. 

(5) In 37, 18, the insertion of ‘aut’ 
before ‘expilatas’ is defended by a learned 
note on the Tacitean use of conjunctions 
after asyndeta. 

(6) In 38, 5, the insertion of ‘in’ before 
‘dicendo’ is an easy and almost necessary 
emendation. 

Special attention should also be given to 
the reasons assigned for the support of 
sundry emendations originated by others, 
as for instance ‘ineunte aetate’ (8, 30), 
‘praecerpta’ (9, 20), the insertion of 
‘causas’ after ‘ privatas’ (10, 37), ‘ tuetur’ 
(11, 16), the bracketing of ‘antiquis’ (15, 
5), ‘Apri mei’ (27, 7), ‘aut eligebatur’ (28, 
16), the addition of ‘periculosa extollant’ 
(37, 38), and ‘maxime’ for ‘maxima’ (38, 
19). 

On the other hand some emendations and 
interpretations should be noted as more 
questionable. Ou some it will be sufficient 
to refer to Mr. Peterson's notes, in which 
the view taken seems preferable; as on 
‘adepturus’ (10, 20), ‘quibus praestant’ 
(13, 14), the reading of ‘quandoque...veniet’ 
as a verse quotation (13, 24), ‘imitatus’ 
(18, 6), ‘atque directa’ (19, 4), ‘insolentia,’ 
for ‘olentia’ (22, 22), ‘sui alienique con- 
temptus’ (29, 9), ‘nostris’ (24, 10), the 
bracketing of ‘autem’ (25, 11), ‘ius huius 
(rather than ‘suae’) civitatis’ (32, 15), 
‘patronis’ for ‘ patronus’ (39, 9), and the 
interpretations of ‘novam negotium’ (3, 
20), ‘civitate,’ taken (as constantly) of a 
separate Gallic canton (7, 3), and ‘ vestrum’ 
(10, 14). Also in 10, 25, Dr. Gudeman's 
bracketing of ‘ad causas’ rests on an anti- 


48 THE CLASSICAL REVIEW. 


thesis which seems hardly intended to be so 
balanced, and on rules of usage of possibly 
later Tacitean development. Again, ‘Porcio 
Catone’ would not be a strict antithesis to 
‘Appium Caecum,’ and if a second name 
were added to one so well known, it would 
probably be the praenomen, as we should 
rather say ‘Marcus Cicero’ than ‘Tullius 
Cicero’; nor again does ‘calescit’ for 
‘clarescit’’ seem required in 36, 2; and the 
received reading in 2, 15 might be defended 
against the insertion of ‘quam’ (after 
Vahlen) by supposing that Aper desired 
that his ability should seem fortified by 


thorough professional study rather than by . 


general literary culture. Also Dr, Gudeman 
will have a large array of opponents to his 
view (on 21, 28), that the verses of Cicero 
‘examined in the light of formal develop- 
ment mark a very noticeable advance upon 


the thought-laden lines of Lucretius or the 
artificial hexameters of Catullus.’ 

Attention should be called in conclusion 
to the very valuable and exhaustive biblio- 
graphy and indexes, the latter contributed 
by Mr. Edmiston and Dr. Muss-Arnolt. 

Also a few misprints may be pointed out : 
sestertia for sestertii (p. 86, 1. 9), divur for 
divam (p. 103, 1. 11), ii. to be inserted 
before 43 (p. 123, 1. 5), Germaniam for 
Germanicum (p. 142, 1. 17), Ann. for H. 
and alia for alii (p. 152, 1, 24), 54 for 51 
(p. 161, 1. 11), omit fatetur (p. 258, 1. 5), 
after ‘and’ insert ‘son’ (p. 264, 1. 25), 
voluerunt for voluere (p. 329, 1. 29). 

It would be a great convenience to readers 
if in future issues the notes were bound 
separately from the introduction and text, 
so as to save the constant trouble of turning 
backwards and forwards. 

H. Furneaux. 





BENNETT'S DIALOGUS OF TACITUS. 


Tacitus, Dialogus de Oratoribus, edited with 
Introduction, Notes, and Indexes by 
CHARLES Epwin Bennett, Professor of the 
Latin Language and Literature in Cornell 
University. Ginn and Company: Boston 
and London. 1894. 


Tus edition, published shortly before that 
of Professor Gudeman, had nevertheless the 
advantage of a previous study of the materials 
collected for that work, and is to a great 
extent based on it, but is obviously from its 
size and price intended for the use of a very 
different class of readers. 

The introduction, in twenty-eight pages, 
covers most of the subjects of the larger work 
and does not appear anywhere to differ from 
its conclusions. It needs only to be said that 
it is throughout concise and clear, and fully 
suitable to the needs of school classes. The 
text and explanatory notes are comprised in 
seventy-seven pages, the differences of 
reading, so far as they are noticed, being 
left to a critical appendix of two pages, and 
an index of proper names and short index 
to the notes complete the volume. 

The text in a large number of places re- 
produces the readings mentioned above as 
originated or adopted by Professor Gudeman, 
but in a still larger number differs from it, 
and notunfrequently approachesthat of Halm, 
but may on the whole be called indepen- 


dent. In apparently two places only does 
Professor Bennett introduce a conjectural 
reading of his own. One is in the corrupt 
words at the end of c. 6, where he reads 
‘quamquam quaedam serantur’ &c., presum- 
ably on the supposition that ‘ quaedam’ had 
been corrupted into ‘quae diu’ and thence 
to ‘alia diu.’ On the whole Gudeman’s sup- 
position, that ‘grata’ had dropped out 
before ‘ gratiora,’ or Nissen’s, adopted by 
Peterson, seem better to bring out the anti- 
thesis. The other is in the beginning of 
c. 38, where he reads ‘ exsistit,’ in place of 
the more generally adopted ‘extiterit,’ the 
latter, though nearer to the MS. text, 
involving asubjunctive with ‘ etsi’ contrary 
to otherwise uniform Tacitean usage. It is 
somewhat a pity that in neither case does he 
give in a note the argument in favour of his 
emendation. 

The commentary is clear and useful as far 
as it goes, but perhaps hardly sufficient for 
the needs of young scholars, who would 
probably need considerable assistance in 
attacking for the first time the difficulties of 
this dialogue. The critical appendix might 
also with advantage have been made fuller, 
so as to indicate more nearly the extent to 
which the MSS. have been departed from. 

Also in an edition intended apparently for 
schoolboys to use in class work, English 
school-masters would certainly in general 





+ lee 


| 











THE CLASSICAL REVIEW. 4Y 


prefer that the notes should be at the end of 
the volume rather than at the foot of the 
page. 

One or two misprints may be noted, as in 
the note on c. 28, 8, ‘avaros producit’ for 


‘ producit avaros’ (in a verse quotation), and 
in the text of c. 30, line 8, the omission of 
‘prius’ (see the note ad locum). 


H. Furneaux. 


BLAYDES’ FRAGMENTS OF THE GREEK TRAGEDIANS. 


Adversaria in Tragicorum Graecorum Frag- 
menta scripsit et collegit Frepericus H. 
M. Braypes, LL.D., Aedis Christi in 
Universitate Oxoniensi quondam alumnus. 


ArTer a lifetime’s devotion to the critical 
study of Greek dramatic poetry Dr. 
Blaydes is anxious to ‘ bind up his sheaves.’ 
All scholars will sympathize with him in 
this desire, and will readily allow the excuse 
he gives for not having fulfilled the good 
resolution which he had formed of revising 
his work before allowing it to appear :— 

‘Has lucubrationes meas _propositum 
mihi erat retractare, ut quae nimium 

festinanter ac parum considerate scripta 
essent castigarem, nonnulla levioris momenti 
materiaeque criticae copiam aliquando 
uberiorem reciderem; sed quominus hoc 
facere possem obstabant senectus iners ac 
valetudo non satis firma.’ (Preface.) 

Yet it is matter for regret that so much 
darnel should be left amongst the grain. 
Por it is this admixture which has hindered 
so materially the acceptance of Dr. Blaydes’ 

labours, at least by his own countrymen. 
Only German perseverance can unweave the 
tangle of things good, bad, indifferent,— 
probable, possible and impossible, which are 
brought together by this author’s ingenuity 
and industry. He seems to be incapable of 
cancelling what he has once set down. 
Else a scholar whose judgment often shows 
itself to be at bottom clear and sound, could 
ever commit to print so many things 
which savour not of ‘iners senectus’ but 
-vather of ‘calida juventas.’ Why, for 
example (p. 108), when he has invented 
_ the quite possible dochmiac line xAve6’, 
 Opare, dedp’ adXdorwp repa hoBepds aiparos, 
d he have allowed to stand the tuneless 
unmetrical first attempt, xAves, dpas viv 
 kAvets viv epyov poBepdv aivaros? (P. 108). 
Why admit so many lines without caesura 
and so many objectionable trisyllabic feet ? 
_ Or why should he at once quote Soph. 0.C. 
481 and deny that Soph. could have used 
peuroa for pdr? (P. 37). 
_ NO. LXXV. VOL. IX. 


Some latent feeling both of the extent 
of corruption and of the uncertainty of 
all conjecture seems to betray the critic 
into the illusion that by heaping together 
many conjectures he can come nearer to the 
truth. The ‘ductus literarum’ is certainly 
not always an infallible guide, but where 
that clue fails, the word selected, to be at 
all probable, should at least be ‘ inevitable’ 
in point of sense. Can this be said, e.g., of 
the following ? 

‘Aesch. /r. 372 (D) (ivnddv TiBacaca 
textovov rovov) 7Bdcaca} 7Bycaca Heath. 
dudiBaca Paleius. Fort. eEyotwca, iiotwoa, 
éexxéaga, 4 POeipaca, aut és te(yiopat’ The 
notion required by the context is that of 
bringing low. Has ypvoaca (transitive) 
never been tried ! 

A volume of Adversaria is at best 
somewhat dreary reading. It is seldom 
that such brilliant sparks shine out amidst 
the endless hammering as Conington’s 


(Aesch. Fr. 98 D) 


GAN "Apns pret 


Led ’ ‘ ~ 
det Ta AGota Tavt’ *aravOilew oTpatov 


(xévra_ tavOporwv MSS.), or Scaliger’s 
aipdpputov Kovpeov in Soph. #r. 132 (D), or 
R. Ellis’ és dp[dropas] in the new fragment 
of Eur, Antiope. 

The mention of Conington reminds me 
that forty years since he set me as an 
exercise to correct Aesch. /r, 238 (D)— 


ddwv tals dyvais mapbevols yapenAtwy 
Néxtpwv acrer pi) BAcuparwv péror Bod). 


My attempt was :— 


*58' *av *ris dyvais rapbévas yapnAwv 
, 7 , 
Néxrpwv *érojun Breupdrov *péror Body ; 


‘Would the eye-glance of maidens un- 
attainted of marriage-rites incline so 
towardly ?’ 

I do not ‘stake my reputation ’ on it, but 
I still think that it bears comparison with 
the suggestions (partly his own) which 

» 


50 THE CLASSICAL REVIEW. 


Mr. Blaydes has thrown together in his 
note. 

In the /ragments, as in the text of the 
plays, some readings have been unduly 
obelized. For example, in the beautiful 
passage from the Danaides (Aesch. Mr. 41 
D), why should the naive naturalism of 
tpoca and éxioe have been questioned ? 
Or why should the pathetic oxymoron 
edipepov kaxov in Soph, Fr. 162 (D) be spoilt 
with Nauck’s all too tame dvypepov ! 

May I observe that at the eleventh hour 
or even later Dr. Blaydes seems to have 


become aware of my edition of Sophocles, 


whether directly or through foreign quota- 
tions does not appear. In his notice of 
Nauck’s edition of 1889 he refers to me 
once or twice, and in the Addenda he 
mentions three suggestions of mine with 
distinct approval. That is some return for 
the pains I spent on the Mragments fifteen 
years ago. But I should have been glad if 
this critic’s attention had been drawn to 
some others of my notes. For instance, I 
cannot but still think that zpds 7a 7Bara, in 
Soph. J’. 86, 7, D, is not only unmetrical 
but flat in meaning, and that ‘ untrodden 
and inaccessible’ or ‘unapproachable’ is 
the sense required. My conjecture kai 
tampoouxta was ‘capped’ by Prof. E. L. 
Lushington with xaémpoorédacra. Iam glad 
to find that in /r. 377 D, the emendation 
ypépa POaver, which I thought of indepen- 
dently in September 1876, was also made by 
Meineke. 

The criticism of the Jragments of 
Euripides is a large subject on which there 
is hardly space to enter here. The work of 
deciphering and emending the recently 
discovered fragments of the Antiope has 
been one of the happiest results of contem- 
porary criticism. Dr. Blaydes’ edition 
nearly corresponds with the revised version 
published by Prof. Mahaffy in the Cunning- 
ham Memoirs after communicating with 
various eminent scholars. It shames me to 
find some of my avowedly rash guesses, 
hazarded before the publication of the 
facsimile, quoted in the notes side by side 
with the deliberate and _ well-advised 
suggestions of such persons as Ellis, Blass 
and Weil. It is something, however, to 
find that I have contributed a morsel here 
and there. Ona few points I have still a 
word to say. 

P. 105, Mragm. A, 1. 4. Dr. Blaydes’ 
objection to fxkra seems hardly justified. 
He says ‘passive poni nequit.’ But ‘it has 


gone so far’ (sc. To mpayya) is not passive. 
The ‘ ductus literarum’ must surely count for 
more in a papyrus than in mediaeval MSS. 
And j#xrat, ‘The matter has been brought,’ 
is hardly possible. 

ll. 8,9. I cannot doubt that Blass and 
others are substantially right in reading 
oT 7...) vov...cTHoat, or something of the 
kind. 

1. 10. The certain correction kai cot pev 
was anticipated by E. L. Lushington. 

1.15. I cannot think ‘that this line is yet 
restored, nor that my former conjecture is 
hopeless : qu. 

TapirOu: mpos a&ypav T evrvxys ely Aysnv } 

Lele Qu: 

tives 0€ valova’ évTds ; €k Totus (waTpas) ; 
onpaval’* (ds) dixa(ida ovvtpopov) 7éTpas. 

P. 106. Fragm. B, 1. 1. Qu. (piacpa 
Tar )pas ? 

P. 107, Fragm. B, 1. 15. 1 am glad te 
find zaiéda Nuxréws approved. 

P. 109, Fragm. C, 1. 38. xvow, Mri 
Starkie’s conjecture, should not be disturbed. 

P. 110,: Fragm. C, 1. 50. Nor should 
épupvat, My. Bury’s reading, be rejected. 

P. 110, Fragm. C, 1. 51. In spite of M. 
Weil’s fine tact, this line seems hardly yet 
restored: ékAurodoa appears to be clearly 
indicated in the facsimile (unless indeed 
the scribe took dévdpy for fem. sing.). 

P. 112, Fragm. C, 1. 70. I cannot see the 
merit of @vyotca. 

The discovery of the Phaéthon, first 
copied by Bekker from the palimpsest 
leaves of the Claromontane codex of St. 
Paul’s Epistles, created once as vivid an 
interest as.is now felt in the papyrus of the 
Antiope. 

P. 165. In Fragm. (81-(N) 1. 
D.)— , 
amavra Tadt’ yOpyo* aKamrvaTws ExEt. 
comes nearer to Bekker’s transcription than 
what is here adopted: and is preferable, I 


Sy Tee E SF 


50 (775 


think, to Munro’s reading mentioned on p, _ 


344. In Fragm. 324, 5 (N) I seem once 
to have thought of épyarys (for #Byrjs). 

On the lesser tragedians I have at present 
nothing to offer. 


I hope Dr. Blaydes will accept of these — 


few notes as a tribute of sincere respect for 
his laborious work, of which the value 


cannot be fully appraised in a hasty review. — 


Such things are— 
od Adyous 
Tyswpev’, GAAL TH Evvovaia wHéov. 


Lewis CAMPBELL, 





PHE 





CLASSICAL REVIEW. 5] 


EARLE’S EDITION OF THE ALUESTIS. 


Morvimer 
Maanillan 


Luripides’ Alcesiis, edited by 
Lamson Eanve. London: 
and Co. 


Dr. Earte has brought to the task of 
editing the Alcestis a competent knowledge 
_ of the literature—especially the later liter- 
- ature—of the criticism of the Greek drama, 

and a trained sense of the force of Greek 

words and Greek constructions. Also he 

has not spared labour. At almost every 

line we have evidence that old and new 
- questions of criticism and interpretation 
_ have been looked at from all sides and 
"decided by a fully instructed and un- 
: biased tribunal. The language of the gram- 
_ matical explanations is possibly more un- 
i conventional—grammar in general is treated 


book, and now and then in the interpre- 
A tation of particular words a boy will tind 
no clue to connect the dictionary meaning 
: with that given in the notes. I refer 
_ under the former head to such notes as: 
: 141 @avotcay, ‘short-hand aorist for per- 
fect,’ 537 ds dpacwy, ‘the so-called future 
participle, 998 Gcoic. dépotws, ‘ cross-cut 
comparison,’ 671 ovdeis BovdAcrar, ‘it helps 
nothing, of course, to write ovd’ eis with 
Porson,’ 229 zeAdooa:, ‘the infin. expresses 
tendency merging in result’; and under the 
latter to (eg.) 403, wizvwv here = werdpevos. 
‘This objection however, if it be one, lies 
mainly against the form, not against the 
matter. Though many of the remarks 
Which follow on individual passages are 
necessarily controversial—for space would 
ail for an enumeration of all the points 
_ on which all readers would be in accord with 
the author—it will be clearly seen from the 
_ hature of the emendations which will be 
‘noticed first, and even from that of the 
Views here combated, that the book is a 
valuable addition to the criticism and inter- 
pretation of the play. 
The following are the emendations sug- 
gested by the editor himself (the brack- 
ted ones are not adopted in the text) :—44 
/ for o°, 64 kAa’on for ravoy, 185 eoryev 
or elyev, 235 a second orévagov introduced 
yetween xAov and ray (not mentioned in the 
otes), 237 yOdvov xara yas for K.y.x., 245 
wpavio. for ovpdvia, 254 f. yp’ éretywv ré 
ets; for t.p.; érelyou, 287 (ovx otv Ge- 
ova, With ovk before éperrapny in v. 285), 
M4 céBov for evav, v. 321 bracketed as 




















spurious, 347 po for pov, 458 Kwxvtoid re 
peOpov for Kaxvrov te peOpwv, 514 (cox for 
cov), 528 y' for 7’, 565 f. cot for tT [Prinz 
reads tw] and aivéres for aivéra, 594 (dpéwv 
suggested to fill the gap after ModAoavéy), 
099 (ev for év), 631 rodrov for réov cov, 
632 rowtrwv for te Tov cav [t wove cov |, 
649 kxarOavetvy for xarGavayv, 808 ye for tm, 
986 (Bporav for dvw), 1045 pi) ene pypvijorxes 
kakov (uy pe pyvioKes B, py pe’ dvaprijorys 
PL), 1123 ri Aevoow; and in 1124 yaixa 
AeEw, 1157 (neOwppicperba for 
perOa). 

The following points in the Critical Notes 
seem to call for some comment ; 47 ‘veprépav 
PL,’ Prinz says Pl, 145 ‘za6o. MSS.’ Prinz 
says P has way, 213 adopting Nauck’s 
reading, which (among other changes) omits 
kakov, EK. says ‘xaxév seems to have come 
from v. 221 where xaxov stands over zdpos 
yap which is much like zopos wae in uncial 
letters.’ It is hard to see how this latter 
fact affects the question. 379 note not so 
full as it should be; see Prinz, 552 the 
correction of the MSS. éevodoyelty was made 
by Stephanus in his 7hesaurus, 795 ‘ heporev 
(sic) B’: what is the force of the sie? 845 
the MSS. reading (jeAdyrerAor) is not given, 
1014 ‘ For defence of v. see Explan. Notes.’ 
No defence is given, on 1063 (discussed on 
p. 67 in the Critical Appendix) I would 
suggest the possibility of reading xat xpos 
mat, on 1071 (also discussed in the Cr. App.) 
is it not possible that the original reading 
Was xpi) 0’ dotis Euppwv Kaptepety Geod docw : 
that firstly éo7¢ was added as an explanation 
over éudpwr, secondly it displaced it, and 
thirdly was altered to @ oi? The iris tori 
which E. adopts and hesitatingly follows 
Prinz in assigning to Monk, is certainly not 
in Monk’s note on the passage. 

In the Explanatory Notes the following 
points may be noticed: 85 (yepa) togypy 
dpoupets érAioas is taken as ‘hold armed 
with the bow in protection (of her)’: this 
is bold and (I think) sound. 50 Bursian’s 
dpBadeiv, which E. rejects, seems to me 
better than the MSS. ¢uBadcv because it 
depends on efoayu, not on rerdypeba: 
moreover v. 527 (réOvny’ 6 peAAwv) shows 
that rots wéAAovor here means ‘those whose 
death is due.’ There is not much sense 
in making Apollo say: ‘no, you are not 
appointed to kill those who ought to die, 
but to inflict death on those whose death is 
due.’ There is no contrast between ‘ those 

E 2 


peeOnppoo- 


52 THE CLASSICAL REVIEW. 


who ought to die’ (ov dv xpy) and ‘those 
whose death is due.’ On 91 f. E. suggests 
the reading, at Soph. 0.7. 80 f., cwrip 
daveins Aapras Gorep Oppacw for cwrnpt Baty 
Aaprpos worep oppatt. 134 rAypes ‘ appar- 
ently the original reading’ [is any other 
reading given anywhere?] in the sense of 
‘in full number,’ ‘as many as were due,’ 
cf. Hec. 522, Hel. 1411 (in both these 
passages the adj. is in the sing.). Is not 
the better interpretation that suggested by 
Hipp. 110 tpdrela zAnpys, Fr. 904 Ovoiav... 
mAnpn? 159 Xevkov ‘not attributive, but 
proleptic with édovcaro.’ 
sight somewhat of a shock, but is well 
supported by E.’s ref. to Hel. 676 ff. 
(Aovtpdv Kat Kpynvav | iva Oeal popdav édai- 
dpuvav). 164 waviorarov ‘neut., agreeing 
predicatively with the following infinitive 
sentence: no comma after airjoopar’: also 
doubtful at first sight, but very possibly 
right, as it makes the change to the direct 
imperat. €vfevfov somewhat less abrupt. 
179 daéAecas KE. (questionably) takes not 
as ‘hast undone’ but as ‘hast lost.’ 304 
If o¢Bwv is accepted the construction may 
be supported by Aesch. Pers. 838 avé&erau 
kAvov, though that passage also suggests 
that the dvadoyou here means not exactly 
(as E.) ‘assume the obligation,’ but ‘be 
content to.’ 320 tpitatov jap is suggested 
as a possible correction of rpityv por pyvos if 
the v. is retained. 371 cionxovoate ‘as one 
might say, have received into your ears’ ; 
is not eicaxovew rather ‘ get to hear,’ ‘ over- 
hear,’ ‘ hear with difficulty’? 572 ‘see note 
on v. 320’: unlike most lines 320 has no 
note. 376 for ‘dactyl’ read anapaest.’ 437 
oixerevers ‘be an oikétis (=dudirodos) re- 
ferring apparently to attendance on Proser- 
pina’; but if oikérus could be used by 
Theocr. (18, 38) for ‘ mistress’ surely oike- 
tevw could have the meaning which the 
dictionaries give it here of ‘inhabit.’ 487 
amremev ‘say no to’: rots movos: ‘ personi- 
fied’: better read zdvovs with Monk, in 
accordance with the usual construction of 
amevretv (cf. Kur. H./’. 1354 Heracles says 
ov [wove] ovr’ damemeiv ovdev Kt.A. 500 
‘ Parenthetical, assigning the reason for the 
preceding exclamation.’ What exclamation ? 
520 The reading in the text is wépu: the 
note presupposes the reading ér. 548 év de 
k\yoate Oipas petavdous: ‘ év=évdov.’ I sus- 
pect we ought to read ed for év here, but if 
ev is retained ev d€ is better taken in the 
ordinary way I think. 595 The scholiast 


This is at first: 


shows that he took Aiyafwv’ as an adj. 
E.’s note is: ‘the sea-giant (cf. A. 403 f.) 
for the sea itself.’ A propos of the scholiasts, 
it seems to me that they might have been 
more often cited with advantage in the notes. 
630 év diAowe ‘neuter’: but cf. 1037, 
640 ‘ds efi=otos ct differing from tis ef as 
qui sis differs from quis sis.’ I do not 
understand. 679 ‘Dobree thought we should 
either omit this vs....? Dobree only proposes 
to omit the p’, not the.vs. 692 f. ‘Cf. fr. 
537’: a better reference would be to J.7’ 
481. 764 réyyovres: ‘sc. dupa, te. while 
we wept’: better (dupa 0 oik edeikvupev Eve 
Téyyovres) ‘ we did not let our guest see that 
we were weeping.’ So at 826 where EH. says 
noOopunv :. ‘felt it, better ynoOounv idov ‘1 
remember noticing,’ lit. ‘I noticed that I 
saw. So too at 1158 where E. says 
dpvjcouar: sc. evtvxeiv, better ‘Iwill not deny — 
that Iam happy.’ 777 E. suggests BeBap- — 
Bapwpévors for the MSS. -vm at Soph. Ant. — 
1001 f. Atv. 779, at 834, and at 1049 the — 
emphasizing kai (od kai ode Oarre 834) is — 
trans. merely by and, too, as well, 795 imep- 
Boddv: ‘the metaphor is taken from — 
throwing overboard of cargo’: rather it 4 
means ‘ get the better of.’ 798: ‘the falling — 
oarage of the wine-cup’ : no need to suppose ~ 
mitvAos was only used of the plash of oars. 
836 ék mpoacriov: ‘with xarowe. ex does 
not = €éw here.’ Is it not better to take it — 
as = év tpoactiw, as it were ‘looking at you 
Jrom the suburb’? 1032 évrvxdvre: ‘to me 
that happened upon (the games’)? ‘upon — 
the prize.’ 1144 kdXvewv ‘answer.’ Ido not — 
understand. Hera is explaining why 
Alcestis does not speak and says ovrw O¢uis 
go. THOSE Tpocdhwvnyatwv Kew mpiv av K.T.A. 
There is a capital Introduction to the 
book, containing, among other things, some 
sound and new arguments in favour of the 
old-fashioned view of the raised stage, to | 
which may be added one urged by A. 
Miiller in the Berl. Phil. Woch. of Nov. 10th — 
in a review of a work of Christ’s (which | 
also takes the old view) that the spectators . 
on the back rows at all events in the | 
theatres of Epidauros and Syracuse could | 
hardly have seen anything of the actors on 
the orchestra level for the bodies in the — 
ranks in front of them. Some notes on 
the Lyric Metres beyond the mere reference — 
to J. H. H. Schmidt would have added to | 
the value of the book. 


¥ 























E. B. Encuanp. 








THE CLASSICAL REVIEW. 53 


GRAVES’S EDITION OF THE PHILOCTETES. 


The Philoctetes of Sophocles, by Frank P. 
Graves, Pu. D., Professor in Tuffs Col- 
lege, Boston: Leach, Shewell, and San- 
born. 1893. 


Ix the preface to his edition Professor 
Graves remarks very justly that the Philoc- 
tetes has not received from American 
scholars that degree of attention that has 
been given to some other plays of Sophocles. 
That this tragedy, in «which the human 
interest is so large, will be brought more 
generally to the attention of American 
students through this serviceable edition is 
to be expected. 

The text is based on that of the 
Schneidewin-Nauck edition of 1887. In 
the treatment of the text, however, the 
editor has often, and mostly with good 
reason, rejected Nauck’s departures from 
the - Laurentian reading. The _ rejected 
readings are placed at the foot of the page 
together with the text of L. Whether this 
was worth the while without indicating also 
the source of those readings adopted by the 
editor that vary from L and from Nauck, 
may be questioned. 

One of the chief merits of the edition 
is a negative but yet a real one—it is not 

 over-edited. Very little material is given 
that is not of immediate service to the 
student, and none that is intended merely 
to show off the supposed learning of the 
editor. But occasionally this merit goes to 
the extreme and becomes a fault. From a 
desire to be brief and concise, difficulties 
are sometimes wholly ignored, or passed 
over so lightly and briefly as to leave the 
impression that the editor has not grasped 
the point or has failed to explain it. Why, 
for instance, has the editor given in v. 782 
the impossible text dAAG dédoix’ & zai, pH pe’ 
_drehijs edyxy without a word of comment, 
except on the use of drehys? In vv. 22 ff. 
he reads oyjpaw’ cir’ éxer xOpov pds adbrov 
_ rovbe y' cir’ GAXy Kupet, which is translated 
by ‘let me know whether he still occupies 


this very spot &c.’ Has the editor taken 
his text and his interpretation from different 
sources } 

The book would be improved if more 
cross references to similar words or usages 
in the text had been given. The passages 
in illustration of the text are usually quoted 
with sufficient fulness, but sometimes only 
the docus is indicated, which in an edition 
for undergraduate students is useless. 

There are occasional slips like the 
following: v. 336, xravov—Oaviv ‘a com- 
mon way of repeating the same verb 
(parechesis).’ 

In v. 733 the editor allows the hiatus 
in ti éorw, but in 753 he writes ri 8 forw. 
The hiatus after 7/ is doubtful in Sophocles, 
and we should probably transpose in v. 100 
and read ri p’ ovy and insert 8 before eras 
in 917, with Jebb. 

The editor has failed to notice the lack of 
metrical responsion in 1118 with 1097 of 
the strophe. The easiest way out of the 
difficulty is to change éuas to duas. On 
perorw V. 1188 the statement is made that 
the only other classical instance of this 
word is in Ap. Rh. But the Greek of 
Apollonius Rhodius cannot be taken as 
a standard of classical usage. In 1213 the 
opt. with wés dy is used to express a wish, 
but is properly a potential optative, not, 
as the note seems to imply, an opt. of 
wishing. 

The edition contains in an appendix a 
brief discussion of the spellings adopted, 
but gives no reason for not adopting »v 
for ev in such augmented forms as é&yiipicxe, 
although this form, according to Meister- 
hans, was in use before 400 B.c. 

A full discussion of the metres, of which 
the schemes are taken from Schmidt, adds 
to the usefulness of the book. 

The publishers have done their part 
satisfactorily, and the editor may be con- 
gratulated on his good proof-reading. 

M. L. D’Ooge. 


University of Michigan. 


: PAGE’S EDITION OF THE AZFNEID. 
* 


The Aeneid of Virgil. Bks. I.—VL., edited 
___ with Introduction and Notes by T. E. 
| Paar, M.A. London: Macmillan & Co, 
1894, 6s, (Classical Series.) 





In Macmillan’s Elementary Classical Series 
has already appeared in separate parts the 
bulk of the notes contained in this volume. 
That is to say, the notes on Bks. iv, and vy. 


54 THE CLASSICAL REVIEW. 


only are quite new ; while the notes on Bks. 
i, li., iil, and vi. have been expanded and 
in many places corrected from the little 
separate editions mentioned above. Those 
who have seen and used Mr. Page’s excellent 
edition of the Acts of the Apostles and 
have had the advantage of his notes on the 
Odes of Horace will know what to expect 
here—good scholarship, clear businesslike 
explanation, no shirking of difficulties, the 
not infrequent clearing up of old ones, and 
withal, the slightest suspicion of the school- 
master in the correction of the .errors of 
eminent scholars. Those who expect these 
things will not be disappointed. 

The introduction which Mr. Page has 
prefixed to his work does not call for 
extended comment. It contains enough to 
enable the schoolboy or passman to take an 
intelligent interest in both poem and poet, 
but does not enter into comparison with the 
late Prof. Nettleship’s little volume on 
Virgil, nor with other well-known intyro- 
ductions and essays. It must be confessed, 
however, that it was a happy thought to 
print Tennyson’s Ode to Virgil, which 
with a poet’s insight goes to the very heart 
of Virgil’s poetry. The text is apparently 
eclectic. But this can only be deduced 
from comparison, as Mr. Page gives no 
indication of source, printing select variants 
at the foot of the page without comment : 
and even in the notes we find too often only 
‘many MSS.,’ or ‘much better MSS. 
authority.’ Whether our surmise is well 
founded we will not decide dogmatically, 
but at any rate we find that in places Mr. 
Page’s text agrees neither with Conington’s, 
nor with Nettleship’s—printed in the new 
Corpus Poetarwm—nor yet with Ribbeck’s. 
To come now to the most important part of 
the volume—the notes: we find them to be 
such as, we said above, we have learned to 
expect from Mr. Page. They are always 
fresh, always instructive, always to the 
point. In more than one place has the 
editor thrown light upon a dark spot and 
has succeeded in explaining difficulties 
which previous editors have either never 
felt, or, not being able to explain, have 
discreetly left alone, or, may be, in at- 
tempting explanation have only added 
clouds to the already existing darkness. 
We would call attention in Bk. i. to the 
notes on 393—396 and 703. In Bk. ii., 
74 is much improved by making everything 
that follows ‘quo sanguine’ oblique. And 
in the same book the explanation given of 
493 was certainly wanted, and we do not 
remember to have seen it elsewhere in 


elementary books. The well-known crux 
iii. 684—7 is carefully handled and, it may 
be said, with more success by Mr. Page than 
by his predecessors. In Bk. iv. 256 Mr. 
Page reads (as does Nettleship) ‘ad Libyae,’ 
which makes good sense and has good MS. 
authority. The note on iv. 459 is new and 
well illustrates the way in which Myr. Page 
makes Virgil his own commentator. In 
the note on 689 we have a protest (which 
appeared first in the pages of the Cl. Rev.) 
against the rendering “grides’ or ‘ grided’ 
for stridit. If the sentence in which 
‘stridit ’ occurs were read in the light of what 


follows, ‘stridit’ would possibly represent 


the sound, untranslatable in English (Lat. 
poppysma), which is made by two wet 
surfaces being forcibly separated as would 
be most likely the case when Dido ter sese 
attollens... | ter revoluta toro est. In Bk. 
v. Mr. Page has hardly succeeded in 
making the Ludus Trojanus clear beyond 
doubt—but will any one? Turning now to 
Bk. vi., Mr. Page’s explanation of 567 castig- 
aique auditque dolos is certainly better than 
the one usually met with. Equally good 
seems to be the explanation of 615 and the 
famous words in 743 quisque suos patimur 
manes. In line 882 Mr. Page adopts 
Wagner’s stopping, by which the passage 
gains both in sense and beauty. 

In more than one place Mr. Page enters 
his protest against the grammatical figure 
tatepov mpotepov, though we think unsuc- 
cessfully. Did he ever put on his ‘coat and 
vest’? The fact is that this particular 
form of expression is a psychological before 
a grammatical question. The mind grasps 
the most impressive fact first. In the note 
on vi. 603, the explanation of ‘fulera’ 
might have been assigned to its author— 
Prof. Anderson of Sheffield. It is hardly 
old enough to be taken into notes without a 
word of its origin beyond a reference to the 
Cl. Rev. 

We have only been able to note a few 
points out of many, but perhaps sufficient 
has been said to indicate the character of 
the book. As a whole, there can be no 
doubt that this edition is far and away the 


best of the smaller ones on Aenezd i.—vi. and — 


in good sense and scholarship inferior to 
none, great or small. It is to be hoped 


a) a 


Qn 


that Mr. Page will find time to give us a : 


complete edition of Virgil with a suitable 


introduction and an adequate index—a 
feature which is wanting in the present 
volume. 

H. ELLersHaw, 





THE CLASSICAL REVIEW, 55 


REICHEL ON THE HOMERIC ARMOUR, 


Veber Homerische Waffen: Archiologische 
Untersuchungen, von Wourcane REIcuen. 
Hilder. Wien, 1894. Pp. 151. 


THE question of the connexion between the 
Homeric poems and the culture of the 
Mycenaean epoch turns mainly, from an 
archaeological point of view, on three pro. 
blems: burial, female dress, and armour. 
The first-of these still awaits a satisfactory 
investigation ; Rohde’s Psyche touches on it, 
but by no means exhaustively or quite 
satisfactorily. Studniczka’s well-known 
treatise deals with the second, and in the 
most vital points does not on the whole 
support the connexion. Reichel has now 
attacked the third with great originality 
and force, and proves convincingly, to me at 
_ least, that here at any rate the Epos is in 
complete accordance with the Mycenaean 
monuments. His views are revolutionary, 
but I think he proves them up to the hilt. 
He deals first, and at greatest length, 
with the shield, in accordance with the over- 
whelming, and at first sight disproportionate, 
importance of this article of the panoply 
in Homer. His point is that the only shield 
known to the old Epos is the Mycenaean as 
we see it in gems and other pictures. It 
was a huge shield, reaching from the neck 
to below the knee, with no handle, but slung 
by the telamon across the left shoulder. It 
was sometimes oblong, but with a curved 
upper edge ; more commonly made of layers 
of bulls’ hide approximately circular in 
shape, but drawn in at the middle so as to 
_ give the characteristic Mycenaean form, 
_ which has so often been treated lately ; see 
_ Mv. E. A. Gardner and Mr. Evans in J.//.S. 
xiii, When we hear of the xv«Aou of the 
shield, it is to these circular layers that 
reference is made. The small circular 
} buckler is entirely unknown to the old Epos. 
| 


ae 497°4= 


mie _ 


It appears in fact to have been introduced 

into Greece only about 700 3.c., when 
_ ‘Homer’ had virtually attained his present 
_ form. So far Reichel’s only novel point is 
the denial of the buckler. But even this 
enables him to draw important and most 
"suggestive conclusions. He points out for 
instance that the two forms practically 
could not exist together; for they imply 
entirely different tactics. It was in fact 
the ponderous shield which prevented any 
possibility of riding on horseback, and 
necessitated the use of chariots to enable 








the bearer to move at all from one part of 
the field to another. 

But a far more vital deduction is that the 
use of the Mycenaean shield is inconsistent 
with that of the metal breastplate. In all 
the many representations of Mycenaean 
fights it is well .known that no Gipyé 
appears ; the shield is truly dudiBpdrn, the 
wearer covers himself with it in a way which 
makes a breastplate a useless encumbrance ; 
or rather, it is ignorance of the breastplate 
which alone can explain the use of such a 
frightfully cumbrous gear as the huge 
shield. The light buckler and the metal 
breastplate came in together. It follows 
then that the Homeric warriors wore no 
metal breastplate, and that all the passages 
where the Owpyé is mentioned are either 
later interpolations, or refer to some other 
sort of armour. We can here only sum- 
marize very briefly the arguments with 
which Reichel supports, in my opinion sue 
cessfully, this somewhat surprising thesis. 

He points out that in any case the use of 
a breastplate is never ascribed consistently 
to any important hero; those who have it 
at one moment, are the next moment with- 
out it. No breastplate is ever mentioned 
in the Odyssey; and—this is yet more 
striking—none occurs in the Doloneia, which, 
more than any part of the two poems, revels 
in minute descriptions of every other sort of 
armour. The panoply is repeatedly re- 
ferred to as consisting of helmet, shield and 
spears, with no allusion to the @dpng. Ares 
himself, who as war-god should wear the 
typical armour, has no breastplate in E 855 
or O 125, Aias never wears one; nor 
Idomeneus nor Sarpedon nor Glaukos, Thus 
the argument from silence is exceedingly 
strong. 

When we come to the passages in which 
a breastplate is mentioned, we find the con- 
clusion confirmed. The only deseription of 
a Oopné which isin any way minute, that 
of the breastplate of Agamemnon at the 
beginning of A, has already been condemned 
by critics as a late interpolation; the 
Gorgon head and the snakes which occur in 
connexion with it are not elements of early 
Greek ornament. But indeed even this 
description is by no means intelligible ; and 
all through the Epos we get no details 
which help us to understand the @opyg, while 
for shield and helmet we have an extraor- 
dinary abundance both of description and 


56 THE CLASSICAL REVIEW. 


of epithet. We have not a word, for in- 
stance, to tell us how the two plates of the 
cuirass were fastened together, though this 
must have been an important matter, re- 
quiring much mechanical ingenuity. 

In two passages at least where a breast- 
plate does play a part, it makes the story 
confused and unintelligible. These episodes 
are the wounding of Menelaos in. A, and of 
Diomedes in E. If from the former we 
expel a single needless line, A 136, kai dia 
Owpykos roAvdadaArov Apypecro, all becomes 
plain. We can understand how-it is that 
the pitpn can be described as epyya xpoos, 
€pkos GkOVTWV, 7 ot wAEloTOV Ep’TO: We see 
why it is that in the two following similar 
passages, 185 ff. and 213 ff., there is no men- 
tion of the @épyé. Similar difficulties arise 
in the case of Diomedes, for E 795 ff. seem 
to exclude the presence of a cuirass, though 
it has been expressly mentioned in 99. 
The other place where the cuirass of Dio- 
medes is mentioned, © 195, has been already 
condemned by criticism. A number of 
lines where the @oépyé is mentioned can be 
cut out without the least damage to the 
sense ; ¢.g. the formal line devrepov ad PwpnKa 
tept oTnbecow edvve, Which occurs four times, 
rather awkwardly introduced between a pev 
and a de. 

We are then justified in concluding that 
the metal cuirass was interpolated into the 
poems ata date after 700 or thereabouts, 
when it had come to be regarded as part of 
the necessary accoutrement of the hoplite. 
But if is not needful to conclude also that 


the word Owpyé is interpolated wherever it 
occurs. As Reichel points out, it has long 
been seen that the verb dwpyocev must 
have a wider sense than ‘to arm with a 
cuirass’; it means ‘to arm’ generally. It 
is a legitimate inference that @dpyé itself 
had originally a wider sense, such as ‘ gear’ 
or ‘armour’ in general; if it already 
existed in this sense in the //iad, it is easy 
to see how it came to be specialized in the 
sense of cuirass, and thus facilitated the 
interpolation of the metal cuirass. In some 
passages it seems to be used of the shield, 
in others perhaps of the pitpy. 

But we must not follow Reichel any 
further ; it will be sufficient to refer to his 
book for the discussion of many interesting 
points which we have left untouched here. 
His explanation of xvypides may be men- 
tioned among these; he shows that they 
were not defensive armour at all, but gaiters, 
which, like so much else, were a necessary 
consequence of the Mycenaean shield ; they — 
were in fact needed to protect the shins in 
walking from the edge of the shield, not to 
turn the enemies’ weapons. Here then the 
tin greaves of Achilles receive a sufficient 
explanation. I note also with much satis- 
faction that the explanation of the dado of 
the helmet is identical with that which I 
gave some years ago in the Journal of 
Hellenic Studies. But the whole work is 
full of suggestion, and marks a most impor- 
tant advance in Homeric archaeology. 

Watrer Lear. 


BUTLER ON THE ODYSSEY. 


LP Origine Siciliana dell’ Odissea. 
Tipografia Donzuso 1893. 
Ancora sull’ Origine Siciliana dell’ Odissea. 

Acireale, Tipografia Donzuso 1894. 


Acireale, 


Tue earlier of these two pamphlets, both of 
which are reprinted from the Rassegna della 
Letteratura Siciliana, contains a repetition 
of Mr. Butler’s assertions concerning the 
connexion of the Odyssey and Trapani. The 
second states and answers four objections 
which have been made to his hypothesis ; 
‘non so altre di queste,’ he adds, ‘e solo ad 
esse rispondo,’ These four objections are : 
(1) that the dialect and the civilization of 
the epic are Ionian, (2) that many ‘critici 
valentissimi’ maintain that it is not a 
homogeneous poem, (3) that no tradition 


connected Sicily and the Odyssey together, 
(4) that there are no traces of advanced 
civilization in Sicily between the neolithic 
age and the Greek colonies. ; 
To the first objection Mr. Butler answers 
that Ionic was the recognized dialect of 
Epic poetry, and used for that purpose by — 
all epic writers, Ionian or not. The state- — 
ment is true enough for later times, though 
there is no proof that a non-Ionian poet — 
would have composed in Ionic at such an 
early date ; still we cannot say that Ionic — 
was not the recognized dialect for Epic all 
over Greece even then. But he is not con- — 
tent with this; he must needs try to show 
that Trapani was an Ionian colony in 
Homeric times. His sheet-anchor is the 
statement of Thucydides (vi. 2) that certain 








THE CLASSICAL REVIEW, 57 


Trojans flying from the Greeks settled about 
Eryx and Segesta and that a band boxéwv 
tov ard Tpoias settled beside them. He will 
have it that ®wxéwy means Phocaeans, while 
rejecting any alteration of the text, a 
touching proof of mental confusion. But 
let it mean Phocaeans ; does Mr. Butler ex- 
pect us to believe that Phocaea was founded 
at the time of the fall of Troy? He does 
not even attempt to obviate the difficulty 
that Thucydides himself says that the 
Chalcidians were the first Greeks to colonize 
Sicily (vi. 3), so that Thucydides himself 
plainly puts no faith in the story about the 
Phocians. Does Mr. Butler wish us also to 
believe in the fugitive Trojans? One is not 
surprised to hear that a/ag is an alteration 
of Pwxatos and that Homer was laughing at 
his own people in the story about Menelaus 
and the dakar. 

The second objection appears no more 
formidable to me than to Mr. Butler, but 
his disposal of it is significant of his single- 
mindedness on the whole question. The 
Odyssey was composed at Trapani ; if it is 
not homogeneous there must have been 
several great poets at Trapani, which is 
incredible. Therefore the Odyssey is 
homogeneous. Q. E. D. 

The other two difficulties are at any rate 
purely negative; they will not appear 
alarming to any critic in themselves. But 
the remarkable thing is that while Mr. 
Butler amuses himself-by overthrowing to 
his satisfaction these ninepins, he does not 
condescend to utter a syllable on any serious 
question connected with the whole matter 
except that noticed above concerning the 
date of the first Greek colonies in Sicily. 
The most awkward point about the Odyssey 
for him is the geographical knowledge dis- 
played in it. The poet plainly knows more 


about the Aegean than about European 
Greece ; he must then have lived to the east, 
not to the west of the latter. Such is the 
natural and well-nigh inevitable conclusion 
to be drawn, and Mr. Butler surely must be 
aware of this, and yet not one syllable does 
he breathe on the subject. His own view 
is exalted by him on the ground that it dis- 
poses of a difficulty ‘ finora insuperabile,’ and 
when we ask what this is we are astonished 
to find that it is the ignorance of the Ionian 
islands displayed by the poet ! The difficulty 
vanishes if the poet lived at Trapani, It 
does indeed, and so it would if he lived in 
London or New York. But what are we to 
think of a man who has the presumption to 
assure us that this is a ‘difficolté finora 
insuperabile’? It was solved centuries 
before it was even stated by the only view 
that will stand criticism or is based upon 
any tradition worth mention, by the positive 
fact that the Homeric poems came to Europe 
from the Ionian colonies in Asia. But on 
Mr. Butler’s view the poet knew the west 
coast of Sicily and the east of the Aegean 
while knowing nothing of the Ionian islands 
or the Peloponnesus. How did this wasteful 
gap in his geography come about ? 

The author may be congratulated on one 
thing at any rate ; he has apparently thrown 
overboard his speculations on the sex of the 
poet or poetess; we hear no more of the 
banter of her father in the person of 
Alcinous or of the ‘sub-clerical Theocly- 
menus.’ Lightened of these strange wares 
his theory may perhaps commend itself 
better to some patriotic Trapanese than it 
does to his ungrateful countrymen, for, as he 
remarks himself, ‘ho tale buona opinione 
dell’ ingegno trapanese che lo credo capace 
di tutto.’ 

Arrnur Piatt. 





HOLM’S GREEK HISTORY, VOL. IV. 


A. Houm.—Griechische Geschichte von threm 
Ursprunge bis zum Untergange der Selbstin- 
digkeit des griechischen Volkes. Vierter 
Band. Die griechisch-makedonische Zeit, 
die Zeit der Kénige und der Biinde vom 
Tode Alexanders bis zur EHinverleitung 
der letzten makedonischen Monarchie in 
das rémische Reich. (Calvary: Berlin. 
1893.) Garces 

A. Hotm.—The History of Greece, translated 
from the German in four volumes, Vol, I. 


Up to the End of the Sixth Century p.c, 
(Maemillan & Co, 1894.) 


Tue first thing one notices in the fourth 
and concluding volume of Holm’s work is 
that the author has broken through a 
chronological barrier which had hitherto 
been recognized. The received view is that 
the correct place for a Greek history to end 
is the Roman conquest in 146, though the 
tempting place for a historian to lay down 


58 THE CLASSICAL REVIEW. 


his pen is the death of Alexander. But 
Holm challenges this view; and with some 
reason. He maintains that the march of 
Greek history does not really end until the 
overthrow of Antony by Caesar; and con- 
sequently his own History comes down to 
the year 30 B.c. He divides the three cen- 
turies between Alexander’s death and 
Antony’s fall into three periods. The /irst 
may be regarded as the purely Hellenic 
periol ; it is marked by a struggle between 
the monarchical principle, which at first 
prevails, and the principle of free republics 
(323-220). The second is distinguished by 
Roman influence acting on the Greek world, 
and Greek reaction against it (220-146) ; the 
third by Oriental influence in the ‘ Graeco- 
Roman’ world, and Roman resistance against 
it (146-30), Thus Holm formulates his 
schematism of this difficult epoch; and 
there is a great deal of truth init. The 
Greek, Roman, and Oriental periods with 
their subdivisions may be regarded as a dis- 
tinct contribution to the philosophy of 
history ; nor can it be denied that the 
volume before us is better rounded off than 
if it terminated with the victory of 
Mummius. But if the philosophy of 
history is satisfied—the influence of Hegel 
one can sometimes detect in Holm’s pages, 
as when he follows spiral lines of develop- 
ment—the practical question arises, whether 
historians of Greece must in future consider 
themselves bound to carry their story down 
to Augustus. Now the Greek world cannot 
be said-to have been laid to rest in the 
bosom of Rome with the catastrophe of 146. 
Though European Greece acquiesced, the 
Hellenic cities of Asia disliked Rome and pre- 
ferred Mithradates ; and the great struggle 
between Mithradates and Rome is certainly 
a part of Greek history. It was a struggle 
for Greece, and it was a struggle in which 
Hellenic cities, Athens herself, took part. 
Therefore the Greek historian on a large 
scale, to whom it is given to thread his way 
through the Diadochi and Epigoni as far as 
146, is bound, as it seems to me, to follow 
Holm and carry his narrative yet further, 
down to the pacification of Caesar after 
Actium. But, I would take occasion to 
remark, this rule does not concern shorter 
histories and handbooks intended for educa- 
tional purposes. I venture to say that such 
works should terminate at the point where 
Greek and Roman history first touch ; from 
that point the thread of the narrative is 
carried on in handbooks on Roman history. 
Thus, for either a big book or a little book, 
146 B.C. is a bad limit ; for the former it is 


it seem interesting to the public. 


too early, for the latter too late. The big 
book should come down to 30, the little 
only to 220. 

It was only to be expected that this 
volume would be ‘ harder reading’ than its 
predecessors ; yet Holm has been wonder- 
fully successful—and this too was only to be 
expected—in investing with interest the 
history of a period which is very generally 
regarded as almost unapproachable by a lay 
reader. While it is a highly interesting 
period to the student who has the time and 
patience to study the authorities, no harder 
problem could be set a writer than to make 
This 
difficulty is due to the intricacy of the 
political relations and the rapid and endless 
shiftings of the situations. Holm has been 
clear in defining the landmarks, has taken 
care to sum up the situation from time to 
time, has been judicious in his omissions, and 
in truth has done his utmost to help the 
reader. But, in some chapters at least, his 
success has been only partial; and this is 
merely a proof that it is hopeless to convert 
the historical material of this period into a 
thoroughly interesting narrative in the com- 
pass of 600 pages. The cross references 
which the author is forced to make in the 
text from one chapter to another are far too 
frequent to be agreeable. The inclusion of 
the history of culture as well as that of 
political events greatly lightens the weight 
of the book. Droysen, in his large work, 
only dealt with the politics, and Mr. Mahaffy 
in his Greek Life and Thought, assuming in 
his readers a knowledge of the outline of 
the political history, dealt with the culture; 
Holm has united both. It is to be noted 
with approbation that he has rejected the 
barbarous word Hellenistic, as to which his 
criticisms (41-44) appear to be unanswerable, 
though Graeco-Macedonian which he would 
substitute is as unsatisfactory as ‘ Graeco- 
Roman,’ which is sometimes used for ‘ By- 
zantine,’ and rather more awkward. 

Holm finds much to controvert and correct 
in the received opinions as to the merits and 
demerits of both the states and the men 
who appear upon his stage; and the con- 
troversial passages of his book deserve 
attention. 

In the first place he accentuates with force 
and considerable justice the contrast between 
Alexander and his Successors. Recognizing 
the great work done by Alexander in ex- 
tending Greek culture, historians have fallen 
into the mistake of ascribing to the 
Diadochi also the same import in the world’s 
history, of regarding them as actuated by 


Bt 
a 
" 





ae 


Pe ee ~~ = -_ wert 









THE CLASSICAL REVIEW. 59 


the same aims, and as the active and suc- 
cessful continuers of their master’s work. 
Holm strictly tests their claims to the 
honour of such a ré/e and finds them wanting. 
‘Die Diadochen waren Egoisten welche ein 
Genie nachiifften, auch iiusserlich. He 
allows that the Seleucids, through numerous 
foundations of Greek cities, worked in some 
measure on Alexander’s lines and toward 
Alexander’s end, but the praise which he 
bestows on them is carefully qualified. He 
allows too that the first Ptolemies were of 
some use through their furtherance of 
learning. But he jealously weighs these 
deserts, and declines to allow that the 
Diadochi can, like Alexander, be justified by 
their works. His view rather seems to be 
that Alexander’s mission was dropped at his 
death, and after a long interval resumed by 
the Romans ; they were the true Diadochi. 
in this spirit Holm writes an epitaph on 
Demosthenes (p. 18): ‘So waren die Miinner 
aus dem Wege geriiumt, welche in den 
Athenern das Gefiihl der Unabhiingigkeit 
und des Hasses gegen Makedonien zu 
entziinden gewobnt waren und die jetzt, wo 
das Makedonische Stammland nicht mehr 
wie unter Philipp und Alexander eine 
Kulturmission erfiillte, eine durchaus edle 
und lobenswerthe Politik verfolgten.’ 

In the second place he combats the view, 
expressed by Droysen and often echoed, 
that the Athenians of the third century 
were sadly degenerated. He holds that the 
Athenian people was ‘ebenso tiichtig’ in 
300 as in 400 B.c., and that there is nothing 
to indicate a degeneration in the following 
century. He protests against the habit of 
measuring kings with one measuring-rod and 
republics with another. He maintains that 
Alexandria had by no means taken the place 
of Athens as a centre of culture, but that, 
on the contrary, the older city was of far 
greater importance in this respect than the 
new. ‘What,’ he asks, ‘are Callimachus 
and Apollonius to the world in comparison 
with Epicurus and Menander?’ What 
indeed? Certainly the observations as to 
the impropriety of drawing conclusions con- 
cerning contemporary morality at Athens 
from the scenes of the New Comedy are 
thoroughly sensible. Were such an argu- 
ment admissible it would tell in his favour 
and point to a distinct improvement in the 
morality of Athenian ladies between the 
ages of Aristophanes and Menander (p. 
202-3). 

In this connexion it may be noticed that 
Holm regards the influence of the second 


Ptolemy at Athens as usually over-estimated. 


He contests the expression of Wilamowitz- 
Millendorf who calls Athens ‘der iiusserste 
Posten der ptolemiiischen Dependenz,’ and 
he thinks that Athens acted spontaneously, 
and not as an instrument of Ptolemy 
Philadelphus, in the Chremonidean War. 
Athens, he is fain to believe, conceived the 
idea of forming a strong league against 
Gonatas. ‘ Wir werden glauben dass Athen 
einen schinen Versuch gemacht hat mit 
griechischer und iigyptischer Hiilfe das Netz 
welches der Kluge Antigonos Gonatas um 
Griechenland geworfen hatte zu zerreissen’ 
(p. 253). Again he observes that the third 
century is marked by a progress in the 
development of the Greek States. Sparta 
and Athens were the only two important 
states in the fifth century; in the fourth 
Thebes ranked with them; but in the third 
we have four political powers, Athens and 
Sparta, as before, and the two federal states, 
in which people who had lingered far behind 
the rest of Greece came at length to the 
front. Had these not at last asserted them- 
selves, Greek history would have been 
incomplete (p. 337). 

In the third place, Holm comes forward 
as a champion of the Romans against ‘the 
prejudice of modern historians’ and seeks 
to show that in the first stage of their 
intervention in Greek affairs their conduct 
was pure and that they did nothing dis- 
honourable in order to further interests of 
their own. He points out that Rome's 
foreign policy at the beginning was totally 
different from her foreign policy towards the 
end of the second century; and Roman 
character had changed in the meantime. He 
repudiates the view that in her first dealings 
with Greece Rome already contemplated the 
ultimate conquest ; and that the proclamation 
of 196 was a conscious step to the consum- 
mation of 146. And of the four actors in 
the events which preceded the fall of the 
Macedonian monarchy—Perseus, Kumenes, 
Rhodes, and Rome—he makes it clear that 
Rome alone can bear the scrutiny of either 
a political or an ethical critic. There is 
assuredly no good reason to contradict by 
an unsupported hypothesis the plain account 
of Polybius which makes Eumenes act as 4 
‘scoundrel ’—-avoupyéraros is the word— 
and the motives of the Pergamene king are 
only too easy to comprehend. 

In such polemic against current views, 
the reader bas to decide, or at least to ask, 
Whether Holm is partial himself or is merely 
redressing the balance. Of the points which 
he has made in his book, he has certainly 
proved some. And he is always sensible, 


60 THE CLASSICAL REVIEW. 


He knows how to ridicule such a worthless 
argument as that the Universal History of 
Pompeius Trogus must have been a mere 
compilation, because he ‘compiled’ a work 
on natural history which had no claim to 
originality. He does not attempt to find 
any clearly determined constitutional office 
in the expressions used by our authorities to 
describe the position of the regent 
Perdiceas. He sees in their true light 
the acts of cities like Athens and 
Rhodes in paying divine honours to 
Antigonus, Demetrius, or Ptolemy. Such 
actions were merely forms, as he well puts 
it (p. 80), of international politeness ; they 
illustrate certain shortcomings, perversities 
perhaps, of Greek religion, but they afford 
no evidence that the cities which paid the 
compliments had forgotten the meaning of 
freedom. A good point is made on p. 335, 
where it is observed that the constitutional 
check which the Achaians had on their 
stratégos was the power of refusing to pass 
supplies, and is thus analogous to the hold 
which our own Parliament for instance has 
on the Government. But this interesting 
observation makes us only the more deplore 
the absence of any full discussion of the 
working of the federal. institutions of 
Achaia. 

In illustration of Holm’s power of bringing 
out characteristic features, a striking sen- 
tence on Antioch may be quoted (p. 142): 
‘Dem Meere nahe und doch keine Seestadt, 
der Wiiste nahe und doch kein nothwendiger 
Anfangspunkt fiir Karavanen, entspricht es 
dem Reiche, dessen Hauptstadt es war und 
das ebenfalls weder rechte Landmacht noch 
rechte Seemacht war; eine kiinstliche 
Schopfung alle Beide.’ In the same way 
nothing could be better than his short de- 
scription of the geography of Asia Minor. 

referring to the passage in the seventeenth 
Idyll of Theocritus, where the dominions of 
Ptolemy Philadelphus are enumerated, Holm 
finds much exaggeration, and observes : ‘ dass 
Ptolemaios Herr der Kykladen gewesen sei 
ist gewiss nicht richtig.’ But is not this too 


SONNENSCHEIN’S 


A Greek Grammar for Schools based on the 
Principles and Requirements of the Gram- 
matical Society. Part I. Accidence. 
Part II. Syntax. By E. A. Sonnen- 
scHEIN, M.A. (Oxon.), Professor of 
Greek and Latin in Mason College, 


strong, and had not Theocritus some reason 
for his flattering statement, in view of the 
close connexion established by Ptolemy 
Sotér between Egypt and the xowov of 
Delos? Holm himself refers to the HroAcuata 
(p. 175), and there is some ground for sup- 
posing that Ptolemy was president of the 
kowov (see Mahatfy, Greek World under 
Roman Sway, p. 108). The words of 
Theocritus should be rather treated as an 
illustration of the discoveries of M. 
Homolle. : 

It is needless to say that Holm is abreast 
of the latest literature in all languages, and 


‘the criticisms of books in his notes are not 


the least valuable feature of his history. He 
must be congratulated on the appearance of 
the first volume of an English translation, 
which Messrs. Macmillan have had the 
enterprise to publish at the extraordinarily 
low price of 6s. net. The English version is 
thus far cheaper than the German original, 
and has also the advantage of having been 
revised by the author in the light of such 
discoveries as have been made (like the 
’AOnvatwy ToAcreva) since 1885. He has not 
of course been able to take advantage of the 
second volume of E. Meyer’s Geschichte des 
Alterthums, or to consider the new character 
which the sixth city of Hissarlik is assuming 
under the divining-rod of Dorpfeld and the 
spades of his workmen. In many ways 
Holm’s first volume is weaker than its suc- 
cessors. He maintains the ‘ Ionian theory’ 
of Curtius; his treatment of the Second 
City of Hissarlik is indefinite and mis- 
leading ; his account of the legislations of 
Solon and Cleisthenes is unsatisfactory ; it 
might be added that he believes in the 
historical personality of Lycurgus. The 
strongest point, perhaps, in this volume is 
his sensible treatment of the Delphic oracle. 
I have noticed a misprint on p. 185, which 
might mislead a beginner: ‘ We have cause 
to regret especially the loss...... of the 
Politics of Aristotle,’ where Politics should 
obviously be Polities. 
J. B. Bury. 


GREEK GRAMMAR. 


Birmingham. (Parallel Grammar Series. 
London: Swan Sonnenschein & Co. 
New York: Macmillan & Co.) 


THE appearance of Prof. Sonnenschein’s 
Syntax brings to a happy conclusion that 


i 





THE CLASSICAL REVIEW 61 


scholar’s labvurs on Greek Grammar. After 
the very flattering reception with which his 
Accidence has been welcomed by the most 
competent judges, it is scarcely necessary to 
add here any words of commendation, 
Suffice it to say that this Accidence com- 
bines completeness and accuracy in such a 
way as to make it indispensable to teacher 
and learner alike. It is eminently up to 
date, gathering together the results arrived 
at by such investigators as Rutherford, 
Meisterhans, Hartel and Kaegi. It possesses 
the rare merit of containing nothing which 
must afterwards be unlearnt. One could 
have wished that the author had more fre- 
quently resorted to the judicious use he has 
made of small type, so as to secure greater 
comprehensiveness, especially in detailing 
exceptions. 

The Syntax marks a new departure from 
accepted routine—a departure moreover 
decidedly for the better. It is based on 
thoroughly scientific principles. Indeed 
too much praise cannot be bestowed on the 
method adopted, and the excellent manner 
in which it has been developed in detail. 
In its main outlines this system has long 
prevailed in continental schools ; but it has 
here been expanded and handled so deftly 
that the author may justly claim it as his 
own. ‘The system of analysis on which the 
whole structure is raised is briefly and 
lucidly set forth in an introduction of five 
pages—-a model of neat and _ precise 
exposition. 

There are two parts. The first treats of 
Sentence Construction on the plan of logical 
analysis applicable to all languages, and 
given in the introduction. To become 
familiar with this system is in itself a valu- 
able acquisition, as it furnishes its possessor 
with ready-made categories by which he 
may readily assimilate any other language. 

Part If. is in a manner supplementary 
and approaches the same matter from the 
point of view of morphology. It groups 
together the various forms which go to 
build up a sentence. The double treatment 
leads necessarily to repetition ; but, as the 
standpoint is different, the only effect will 
be a deeper ingrafting of the whole on the 
mind of the learner. 

As this is a book which is sure to go 
through more than one edition, and as its 
framework is a xrijpa és deé admitting of 
improvement without substantial alteration, 
it will not be amiss to offer a few comments 
pointing out omissions and suggesting 
emendations. 

To begin with the more crucial points of 





sentence construction, namely pages 179 
sqq. Which deal with the moods of the 
simple and complex sentence. In 34la, 
obs. 1, 3414, obs. 1, Prof. Sonnenschein has 
retained Curtius’ well-known distinction 
between present and aorist jussives. Com- 
mands and prohibitions applicable to a single 
occasion are said to affect the aorist, and 
such as are applicable to general rules of 
life, the present.!. It is somewhat strange 
that it is not even hinted that the instances 
where this law is unobserved by Greek 
writers are so numerous, that the rule itself 
must either be abandoned or in some way 
patched up so as to meet the countless ex- 
ceptions which may be arrayed against it. 
In dealing with other modes of rendering 
command (341la, obs. 3) it would be advis- 
able to add another illustration of coupled 
command and prohibition such as Eur. //e/. 
436 odx dradAdfea dépwv kati i)... dxAov 
mapéfes; It should also have been stated 
that the particle od is sometimes omitted 
e.g. Androm. 253, Ar. Pax 259, Vesp. 671. 
The third mode of command should include 
an example of the third person of the 
indicative, ¢.g. 


? > , ‘ , ; 
dus éxeivw TH Adyw pabyoerat. 


Ar. Nub. 882. 
‘The same remarks applies to prohibitions: 


Hevbeds drws pip 7évO0s eicoloe doj.015. 
Bacch. 367. 


An addition to 341) in the shape of an 
obs. 3 is desirable to set forth the other 
modes of rendering prohibition. At least a 
reference should here be made to the por- 
tions of the book which treat of the pro- 
hibitive force of od py and éxws py with 
future indicative. Instances might with 
advantage be added illustrative of the 
particles employed when another prohibition 
or positive command follows on od py, ¢.9- 


> ud AaAnjoeis GAN dxodrovbsjoas ; 
sg ; Ar. Nubd. 508. 
od pul rpocoiaes XEtpa Bayxetoas 8 iov 
8’ eLoudpter pwpiay tiv ory €pob ; 
aa Eur. Bacch. 343. 


In dealing with wishes, our author is, in 
my opinion, justified in implicitly admitting 


: ioe pe as a 
1 This is the corollary or practical application of 
outa principle which is also given and oe 
with the Editor’s kind permission, I hope to discuss 
on some future occasion. 


62 THE CLASSICAL REVIEW. 


ds &pedov as a possible prose idiom to mark 
a wish -unfulfilled in the past. The use of 
ris av similar to that of was av is so rare 


that it is as well not mentioned. Under 
exclamations (345) there is a strange 


omission of the Greek idiom corresponding 
to the Latin coniunctivus indignantis, e.g. 
ue wabeiv Tade hed. The index refers us to 
other portions of the work where this 
subject is treated ; yet in none of the places 
so indicated is this exclamational idiom to 
be found. It is however not altogether 
forgotten, but certainly here is its proper 
place. 

Temporal, Local, Comparative or Modal, 
and Adjectival Relative Clauses are appro- 
priately treated on uniform lines ; they may 
be here conveniently discussed together. 
The twofold classification of actions marked 
as facts and actions marked as prospective or 
general has the obvious merit of brevity, but 
is lacking in precision and comprehensive- 
ness, at least as set forth by the author. 
In the first place the disjunctive epithet 
‘prospective or general’ fails to bring out 
the common element evidently underlying 
such forms as drav, d7ov av, ws av, Os ar. 
If we compare the two types ov ép@ and oy 
av dpe, it will be readily ohserved that the 
former, besides marking a fact, also marks 
the action as definite and particular, refer- 
ring as it does to definite and particular 
persons. The latter on the contrary is 
always in some way indefinite. This remark 
also applies to local, temporal and compara- 
tive clauses. Our meaning will be better 
illustrated by a tabular statement :— 

I. 6 dvjp dv dps. 

If. (1) darts dv 7) 6 avijp Ov op® eb otd" drt 
Kaos eote Kayabos. 

(2) Ov dv 6p TotiTo dpOvta wodp’ eras. 

(3) Ov dv OpO araxtws idvta KoAdoopat. 

Class I. expresses definite fact. 

Class II. (1) 6ot1s av 7» is merely in- 
definite. 

(2) ov av bp& is indefinite and iterative. 
The method of description is non-particu- 
larization and expression is given to a 
sequence of habitual actions. 

Class IT. (3) is indefinite and prospective. 
The time referred to is the indefinite future 
and the primary notion conveyed is that of 
a future contingency. 

Prof. Sonnenschein’s term ‘ general or 
prospective’ practically covers Class II., but 
fails in not giving the genws. Moreover he 
has not defined the word ‘general,’ being 
content to describe it as ‘ ever-clauses.’ 
Now no two English writers agree in their 
practice of inserting or omitting the affix 


‘ever, so that at best this must be con- 
sidered a somewhat shifting criterion. 
Indeed it may be taken as a safe maxim 
that English word-equivalents should only 
be adopted as aids to grammatical exposi- 
tion by way of illustration from the verna- 
cular; they should not be employed as 
substitutes for grammatical enunciations, 
still less as definitions. Further, the term 
‘iterative’ or ‘frequentative’ conveys a 
valuable syntactical notion that has to be 
used elsewhere, so that it ought not to be 
set aside here. Prof. Sonnenschein has 
also, it appears, omitted to emphasize! a 


‘special form of Class IT. (1) above mentioned, 


which is commonly called ‘ generic’ and 
deseribes an undefined class as opposed to 
particular actions or individuals, e.g. ots av 
 Tavtov atpa=persons of the same blood. 
The corresponding idiom to mark a class 
described indefinitely and negatively, for 
which p7 with indicative is the prevailing 
form, has not been set forth, cf. ui wet?’ & 
py Ot, 7.€. TH py Tpoonkovta. If it be added 
that, in the use of affirmative generic rela- 
tive clauses, the Greeks seem to have 
selected the particularizing or non-particu- 
larizing mode of description according to 
pleasure, we have all the facts connected 
with this construction. See R. Whitelaw’s 
note in Classical Review, April 1894, and 
compare the two following passages: (1) 
dspoow Omep ot Sporns Oaor av O€woury 
ed do tav Kdtw, azo b€ Tov avw py. (2) 
TavTOV OG EaT Euvxa Kal yvounv ExEL.— 
Med. 230. 

The ‘special rule for zpiv’ (347, 3) is that 
found in most grammars adapted to the 
terminology of this book. It represents 
the orthodox? view, and was learnt, held and 
taught by the present writer till he came 
across passages which weaken two of the 


1 This idiom has not been overlooked as, I think, 
I have met it somewhere, though I cannot now 
recover the passage. It does not occur in the places 
referred to by the index as ‘ Generic.’ 

2 Mr. Marindin has been good enough to point out 
that the strictly limiting doctrine stated above as 
orthodox scarcely deserves that name. He refers, 
among older grammars, to an appendix to Madvig’s 
treek Syntax recording the occurrence of mpfy with 
ind. after affirmative clauses. This use is there 
described as ‘ccmparatively rare and mostly 
confined to Tragedy and Thucyd.’ He gives a 
further reference to Donaldson, who simply notes 
the use of the indicative after affirmatives and 
negatives. Hence, as Mr. Marindin suggests, the 
rigidly restrictive rule should be described rather as 
a new heresy than as old orthodoxy. Nevertheless 
Prof. Sonnenschein’s version of the rule is very 
common in recent grammars and is constantly 
repeated by commentators. It is needless to add we 
are dealing exclusively with the Attic use of mptv. 








aa ed ea 









































assumptions on which it is based. ‘hese 
are: (1) the exclusive occurrence of the 
indicative with azpiv after a negative 
principal clause: (2) the restriction of zpiv 
and an infinitive to cases where the principal 
clause is affirmative. 

Contrary to No. 1 we find the indicative 
following on an aflirmative— 

oi &. tporeuuryov tots ’A@nvatos Kai dvre- 
KATAOTAVTES Tals Vavol TOV aiTov TpoToVv abéis 
dujyov mpiv by Apiorwy wecGe.—Thue. 7, 39. 
(Arnold’s text. ) 

wapatAnow S€ Kal of éxt Tov veov adrois 
émacxov, mpiv ye 51 of &. érpeway te Kal... 
katedtwxov.— 1b. 7, 71,5. (Arnold places a 
4 colon after éracyxor.) 


} \ be > ¥ \ s / 
TTOVOL O€...yTaV Lat TwS Tp O A. TeLHeL. 
Kur. Hee. 133, 
avodoAvke mpiv y dpa. 


Eur, Med. 1173 (Paley). 


The two latter instances are given by 
Passow in his Wérterbuch. He there tells 
us that ‘die construction von zpiv mit dem 
Indic. findet statt eben sowohl nach posi- 
tiven als nach negativen Hauptsiitzen.’ 

Contrary to No. 2— 

avaBiBalew Srws pi) TpoTepov VE erat Tp 
muléoGat Tovs avdpas.—Andok. Myst. 43. 

od Bovidpevos payn diaywvicacGar zpiv ot 
Kal tovs Ponfodts HKew. . . exéAcvev.—Thue. 


v. 10. 


4 ‘ a a > a / 
TO yap ToUotv exaoTos éxpabeiv OeAwy 
3 a“ ‘ 2 4 , 
ovk dv peOeiro zpiv Kab’ Hdovav KAvew. 
Trach, 196. 
mod 6€ ony 
py mpiv Tapaéys piv 700’ ed GéoOa, Tekvov. 


Eur. H.F. 605. 


Gporav pay mpi new mpiv i Tov pvdpov 
_ Tovtov avadjvar (avapavivatt).—Hat. 1, 165. 

The last two examples are given by 
Passow, who declares that ‘in Prosa findet 
sich zpiv mit dem Inf. nach einem negativen 
Hauptsatze nur selten.’ 

The foregoing are sufficient to warn us 
against sweeping assertions concerning zpvv. 
wording grammatical rules it is always 
advisable to insert those valuable safe- 
guards ‘usually,’ ‘ generally.” 

1 Since writing the foregoing I have found this 


‘subject fully treated in Goodwin J/, 7’. (ed. 1889) 627 
nd The instances cited above are there given, save 


tives—‘ when piv means simply Jefore and not until.’ 
oa sis of this verbal distinction would 


THE CLASSICAL REVIEW, 63 


Looked at in the light of the rigid canons 
of logic the rules for Hypothetical Clauses 
(353) are tenable, and, with the single 
exception of the statement that the  r 
exo. protasis is exclusively future, they are 
in accordance with the ascertained facts of 
the language. Nevertheless they are open 
to grave objections. It is not enough that 
they should state the truth; they must 
state the whole truth and that with all 
possible lucidity, so as to leave no room for 
misconception, especially to the school-boy 
mind. These latter qualities are in my 
opinion lacking, and I do not hesitate to 
pronounce them insufticient and inadequate. 

The division adopted rests on a twofold 
basis, namely the presence or absence of 
‘what would be’ in the principal clause, and 
the implication conveyed as to the fulfilment 


or non-fulfilment of the condition. In the 
first place the terminology is unphilo- 


sophical and seems out of place in a scien- 
tific work such as this is. The adoption of 
one form of protasis in the vernacular, to 
convey a grammatical notion applicable to 
all languages and of very wide extension, 
is bound to be misleading ; and this is all 
the more unpardonable as the book is also 
intended as an aid to composition. Suppose 
a boy had to put into Greek the sentence : 
‘If it were desirable we might give proof.’ 
As becomes an intelligent pupil he would 
turn at once for guidance to his rule, and 
so would look out for the magical ‘ would 
be.’ Here it happens to be conspicuous by 
its absence; hence he could hardly be 
blamed for arriving at the erroneous con- 
clusion that the sentence in question cannot 
come under Class B. If it be said that he 
ought not to be so stupid as to fail to see 
that ‘might give’ is equivalent to ‘could 
give’ or ‘would be able to give,’ one may 
easily reply that this is only training him 
to interpret words by words, instead of 
supplying him with a criterion in the shape 
of an idea, Even the more backward boys 
will make fewer mistakes when working 
from an idea than in following anything 
even savouring of ‘rule of thumb.’ Again, 
take another boy, whom we shall suppose 
this time to have become quite an adept at 
recognizing ‘would be.’ Put before him 
the sentence ‘were it true, it would be 
pleasant.’ Seeing his talisman he would 
tell you straight off, without thinking, this 
sentence must be marked Class B. But is 
it not to be feared that, having contracted 
the habit of looking for light to the English 
phraseology, he would, in settling his 
further account with ‘if it were,’ jump at 


64 THE CLASSICAL REVIEW. 


the conclusion that this is a case of a con- 
dition stated as unfulfilled, inasmuch as 
more frequently than not this is the im- 
plication of that varying idiom? Here 
however he would be too hasty and might 
fall into a trap, for our word ‘were’ in 
such constructions may be equivalent 
either to ‘if it was’ or ‘if it should be’ (a 
merely conceivable supposition)—according 
to emphasis and context. What is wanted 
is not that there should be a search for 
‘would be’ or ‘should be’ or ‘might be’ or 
other variations; but recourse should at 
once be had to the universal canon of 
fulfilment or non-fulfilment. 

These objections, though not unimportant 
in the eyes of the teacher, may be set 
aside by the scholar as trivial and super- 
ficial. There are other deficiencies of a 
graver sort. Class B is made to contain 
two sorts of conditional sentence so unlike 
in form and meaning as scarcely to hang 
together by a hair. The connexion consists 
in this that each has an implication ; yet 
what an implication? So diverse that it 
might be said with equal truth that the one 
has no implication as to fulfilment or the 
reverse, and the other has the very definite 
implication of non-fulfilment. These two 
subdivisions stand at such a distance from 
each other—albeit they have an implication 
—that they ought to be placed under two 
distinct headings, or else a_ tripartite 
division of the whole subject should have 
been adopted. Merely to mark off the con- 
dition unfulfilled in the past or present as a 
‘special form’ of the class to which the 
colourless supposition belongs, is but a poor 
makeshift and altogether insuflicient to 
characterize their differences, which, as has 
been shown, are far more conspicuous than 
the slender bond of union holding them 
together. 

A more serious defect still—for so far our 
author is within his strictly logical rights— 
is to be found in the exposition of the rule 
regulating unfulfilled conditions. After 
the campaign recently conducted in the 
pages of this Jteview against the very 
principle of fulfilment as a basis of classi- 
fication, one point is now clear, if it was 
uot so already, namely, that to be regarded 
as unfulfilled, a condition need not necess- 
arily be so actually, but that it is enough 
it should be asswmed to be such. Mr. Son- 
nenschein has overlooked this well-estab- 
lished fact, and, besides neglecting to use 
words to bring it out, has even employed 
language which is positively erroneous, 
‘In these sentences,’ he writes, ‘the past 


(Latin) subjunctives or the (Greek) indica- 
tives of the If-clause refer to present time; 
instead of denoting what was they have 
come to denote what is not’ (§ 353). 
Instead of ‘is not’ we should have ‘is not 
or at least 7s asswmed not to be.’ The above 
statement is elsewhere repeated (472, 4) 
and is altogether false as applied either to the 
Latin or Greek idiom. A good illustration 
occurs in Cicero 7.D. 4, 35: Etenim si 
naturalis amor esseé ot amarent omnes et 
semper amarent, where‘the supposition is 
only contrary to assumed not actual fact. 
Love certainly 7s natural, and the speaker 


~might have been fully convinced of this, 


though he chose to assume the opposite. 
Similar instances might be multiplied. 
Hence we must speak of conditions stated 
as unfulfilled, stated as colourless, &e. 

Lastly, the assertion that colourless or 
merely imaginary suppositions of the form 
el Te €xou Ooty av refer exclusively to future 
time is, In my opinion, erroneous. An 
examination of some examples would soon 
bring home the conviction that the essential 
element in these forms is their colourless- 
ness as to realization or non-realization, and 
that the time is merely a matter of infer- 
ence. In nine cases out of ten this may be 
the future, but the tenth case does appear 
where it is the time of speaking, ¢.e. the 
immediate present, or the general present 
for suppositions applicable to all times. 
This will become clear from the corre- 
sponding Latin idiom which has the 
advantage of employing two well-defined 
tense-forms—the present and perfect sub- 
junctive—according as the ideal action is 
put forward as going on or completed. Cie. 
De Of. wi: 6,-29 *Quid? si: Phalaron 
crudelem tyrannum vir bonus, ne ipse 
frigore conficiatur, vestitu spoliare possit, 
nonne faciat?’ Phalaris had died some two 
hundred years before the Stoic wise man 
was heard of. The supposition is purely 
imaginary, 7.e. it prescinds from fulfilment 
or the reverse, and the time is not fixed. 

If we regard Phalaris and the Stoic as 
types, then we have a supposition true at 
all times, ¢.e. realizable now or at any time. 

Cic. 7.D. 2, 4 Si grammaticum se pro- 
fessus quispiam barbare Joguatur; aut si 
absurde is canat, qui se habere velit musi- 
cum, hoc turpior si¢ quod in eo ipso peccet, 
cuius profitetur scientiam. 

Here the actions are represented as in 
progress and the time is left to the reader 
to imagine ; it is only a secondary considera- 
tion. True, the time of loguatur is rela- 
tively future to that of the profession of 





¥ 
: 


bua 76 py GElav etvar. 
without os or are. 
_ such equivalents is an invaluable aid to 
composition, and indeed the general rule, 
under the system here followed, can scarcely 
_ be said to be complete without it. 
Query : may the conjunctions ézei/, éeidy), 
Ott, di071, &e., be used in these clauses indis- 
_ criminately 4 
attempt at discrimination 4 

___ In dealing with subordinate Concessives 








Beg ve 


common way of expressing concession. 


THE CLASSICAL REVIEW, 65 


grammatical ability, and sit is relatively 
future to Joguatur. No more can be fixed 
about the time indicated, which, as_ the 
supposition is for all times and_ purely 
imaginary, scarcely enters into the thoughts 
of the writer. One other example will 
clear up what remains obscure. Cic. De 
Of. iii. 24, 92 Si quis medicamentum 
cuipiam dederit ad aquam intercutem, pepi- 
geritque si eo medicamento sanus factus 
esset, ne ullo postea uteretur; si eo medi- 
camento sanus factus sit et annis aliquot 
post énciderit in eumdem morbum nee ab eo 
quicum pepigerat impetret ut iterum eo 
liceat uti, quid faciendum sit. Here we see 
the real force of the tenses in this form of 
conditional sentence. The perfects repre- 
sent the action as completed, the present as 
going on. The time at which the supposition 
is supposed to take place is altogether left 
out of consideration. The whole case is 
fictitious and may be referred to any time 
present or future ; the incidents described as 
completed might have reached that stage at 
the moment of speaking. The allegation 
that the forms in -erit may be future 
perfects is precluded by jfactus sit and 
pepigerat. 

The theory of Causal Clauses (349) is 
expounded with conciseness and accuracy— 
which are throughout the characteristics of 
this book. Unfortunately there is no 
observation giving a brief summary of 
Causal Equivalents, e.g. (1) Relative és or 
éotis With indicative (occurring further on 
364, 2). (2) dua and xara with accusative, 


_éxi with dative = on the score of, because 


of, also did with infinitive: airiy drjAacav 
(5) Participle with or 
A short synopsis of 


If not, why have we no 


358) our author has done well in not 
adopting Kriiger’s distinction between xai 
«i and «i xai which, as another learned 
German observes, is ‘nicht immer beachtet, 
Wie ja auch unser “sogar in dem Falle”’ 
jeselbe Bedeutung hat wie “in dem Falle 
gar.” ’ On the other hand an observation 
wanting on «ay «i and the fact should not 
been overlooked that a participial 
phrase even without xairep is not an un- 


NO, LXXV. VOL. IX. 


The. 


use of duws for xairep is so rare in Attic 
that it is as well omitted; duws wai with 
participle is more frequent. 

The rules for Purpose Clauses sin chiefly 
by omission. To the list of Final Equiva- 
lents given by the author (351) add (1) 
such constantly recurring phrases as é7i 
muoree and the less frequent form xara 
mvotw, both expressing intention (literally 
‘with a view to’—‘in search of’ informa- 
tion), (2) the future participle accompanied 
by the article, ¢.g. ot« jv 6 kwAvowv. Obser- 
vations are also missing on (1) the occasional 
use of dws with the future indicative in 
clauses strictly final, i.e. of adverbial sub- 
ordination ; ef. Cho. 257 ovya6’ érws pi) 
mevoetat, Andoc. Myst. 43 dvaBiBdlew drws 
py mpotepov vvé éorac zpiv...(2) on the 
virtually final import of « wws meaning 
‘in the hope that,’ ‘to see if,’ ‘in order that 
perchance’ and common in Attic prose. 
Mr. Sonnenschein has also omitted to notice 
the apparently final use of ézws with poten- 
tial optative as seen in Ag. 364 drws dv... 
BeXos HAvov oxyWeev, Thue. vii. 65, drws av 
arokuaOdvor Kat pa Exor dvtAaBiy ta éuBad- 
Aopeva. A similar omission is also observ- 
able in his treatment of object noun-clauses 
dependent on verbs of effort, where the 
analogous construction oxo7eiy érws dv 
duayouev has been passed over in silence. An 
explanation however of this probably wilful 
omission is to be found in the fact that our 
author is very likely of opinion that this 
point has not as yet been sufficiently 
thrashed out by those scholars who have 
touched on it to be ripe for incorporation 
into a text-book. It would seem on the 
other hand that some attempt ought to have 
been made to distinguish, if not dogmati- 
cally at least provisionally, between the use 
of ws, 67ws and ds av, d7ws av in final clauses. 
German scholars generally speak of the 
latter as unvollstiindige! Absichtsitze in 
contrast to  vollstiindige Absichtsitze. 
Among our own grammarians they are 
spoken of respectively as final and quasi- 
final or semi-final. Hermann maintained 
that és dv is equivalent to dummodo, 
si modo; Paley that ‘it expresses result 
rather than intention.’ Others say that 
és dv represents the purpose as _ con- 
ditional, and their view is, in my opinion, 
nearest the mark. The distinction, if such 
there be, must be drawn from the general 
influence of dy in all those clauses where it 
occurs with the subjunctive (Prof. Gardner 
Hale’s ‘ prospectives’), which the author 

1 This word has different extension with different 


writers, 
F 


66 THE CLASSICAL REVIEW. 


under review explains as ‘prospective or 
general.’ The nuance conveyed by av in 
this conjuncture seems to have the effect of 
rendering the clause in which it appears in 
some manner indefinite and subject to 
conditions. Hence in final clauses it would 
seem to have the force of representing the 
purpose as indefinite, conditional, put for- 
ward with misgiving, or reserve or modesty 
or without assurance. Thus ézws av is 
sometimes equivalent to ‘in order that 
perhaps. Be this as it may, whatever 
distinction applies to strictly, adverbial 
clauses of purpose must, it would seem, 


also prevail in the object noun-clauses with 


subjunctive following on verbs of effort 
(értpedcicbar dws), since the latter have 
borrowed or may be treated as having 
borrowed the final apparatus, subjunctive 
with ozws and orws dv. Dr. Fox has an 
excellent note dealing with this very dis- 
tinction between dzws and drws av in semi- 
final clauses. It will repay quotation here, 
as it helps to throw light on this interesting 
problem :— 

‘ Beide Formen [dézws und ézus av] kommen 
mit Konj. verbunden im unvollstiindigen 
Absichtsatze bei den attischen Rednern vor, 
aber ows av noch viel seltener als ézus, 
indem jenes nach Weber nur je einmal bei 
Ps. Lys. 6, 4, bei Is. 7, 30 und bei D. 19, 
299 vorkommt, wahrend dézws sich bei 
mehreren Rednern vereinzelt findet, bei D. 
jedoch (wie bei den Tragikern, bei Thuk,, 
Xen. und Plat. und den spiiteren Inschriften) 
ofters. Bei drws ay wiirde mehr das Objekt, 
die von Bedingungen abhingige Art und 
Weise des cvuprparrew hewortreten, wabrend 
beim einfachen 6rws der Begriff der Absicht 
stirker, wenn auch nicht so rein wie durch 
iva hervorgehoben wird.’ 1 

In 369a, obs. 3, one of the instances of 
Effort Clauses with dézws dy and subj. is 
oKoTe orws av dmobdvepev avopikwtata: iS 
not this reading of Aristoph. Zg. 81, to say 
the least, doubtful ? 

Under Consecutive Clauses (352) an 
observation similar to 351 is missing to 
record other Consecutive Equivalents. Such 
are (1) oios, doos with ind. and inf., (2) 75 and 
To py With infinitive: pdBos dv6 barvov rapa- 
orate 70 pi) PA€hapa gvpPBadety and contrast 
tov and rod py, (3) ds and doris with indicative 
(given further on 364, 2,c). A record is 
likewise here wanted of the other idiomatic 
uses of gore. The index refers us only to 
368g, where the loose or superfluous insertion 

1 Demosthenes’ Rede far di ypoliten vo 
Wilhelm Fox, S.J. (Heatburgish- Becomes tT 
page 118. 


of this particle after such impersonals as 
ovveBy &e. is chronicled. But there are 
others, notably (1) its frequent substitution 
for éf’ 0, ef wre to express ‘on condition 
that,’ (2) for 6. 6 or roivyy at the beginning 
of a sentence. The idiom petfov 7 adore is 
elsewhere given. Further the frequent 
recurrence of the potential optative after 
gore to express a possible or conceivable 
consequence, as well as that of the past 
indicative and infinitive with dy to express 
an unfulfilled consequence, is so marked as 
to call for special mention even in a school 
manual. This omission should be supplied 
in a second edition. Lastly to bring out 
the contrast between an actual consequence 
and that which some grammarians call a 
natural consequence, in lieu of the latter 
our author adopts the term ‘ contemplated — 
or in prospect.’ Now our word ‘contem- ‘ 
plated’ can, as far as I am aware, have only 
two meanings: existing in thought or in ~ 
prospect, and, as the latter idea is expressly — 
added in the text, it may be assumed that 
the former is the only one intended to be 
conveyed. It however is insufficient, and 
some such word as ‘congruous’ or ‘natural’ 
or ‘to be expected’ might with advantage 
be appended, so as to have ‘ contemplated as _ 
congruous &e.’ 
The Noun Clauses are admirably drawn _ 
up, the subdivisions exquisite, and this — 
treatise might be pronounced perfect, if 
only philosophical exposition was even ~ 
allowed precedence over ‘rule of thumb.’ — 
How much more satisfactory is the heading 
‘dependent or indirect statement following 
on verba sent. et declar., verbs of effort, 
verbs of emotion,’ than ‘ that-clauses’ which 
‘express that something is or should be’ &e. 
Although I fancy we are not over 
particular about the that of that-clauses, 
still if, for any reason, the term has to be © 
employed, let it take second place andl 
appear as an aid to scientific enunciation. 
The several species of the Noun Clause are 
however beautifully worked out in detail, so _ 
that one can pardon the obtrusion of © 
English morphological equivalents and the 
omission of the useful class of verbs of 
emotion which would include (1) con- 
structions with Oavydlo, dyavaxto &e., (2) 
poBotpa wy. The regular idiom after such 
verbs as avtiWéyew &e., with the so-called 
superfluous negative, is alone recorded 368b. — 
It would perhaps be well also to point out 
the other method of simply giving the - 
opponent’s assertion, e.g. dru dpotos €¢ TOVTOLS, 
ovd adtos dv audio Bytnoeas Pl. Symp. 215B. 
A similar remark applies to the analogous — 


oY 









yg se 

















saat 


THE CLASSICAL REVIEW. G7 


construction of ya od with inf., e.g. od8 dv 
els dvteirot pr) ob (Or pr) Tupdhéeperr. 

The treatment of prepositions is alto- 
gether too meagre even for a_ school 
grammar. The general scheme: ‘'Time— 
Place—Other meanings’ is inadequate, and 
gives undue prominence to time and place 
as compared with the other meanings of 
equal or greater importance. For instance 
the all-important instrumental force of dud 
with gen. comes under the heading ‘other 
meanings, placed side by side with the 
adverbial expressions dus yepdv exew, dd 
taxous, instead of forming a special class, 
‘instrumentality,’ which is undoubtedly the 
most common meaning of da with gen. 
All the numbers 448—461 will simply have 
to be rewritten. 

The various idiomatic usages connected 
with the pronouns 06é¢, otros, éxeivos and the 
oblique cases of airds are practically not 


dealt with. On turning to the index at the 
word éxeivos to see if it be elsewhere dis- 
cussed, one is referred to § 567 where this 
pronoun is not even mentioned. 

Space will not allow of a detailed exam- 
ination of other portions of this valuable 
work. The reviewer cannot do better than 
conclude by stating the general impression 
its perusal has left on his mind. He feels 
the more entitled to do so as he has not 
been sparing in pointing out shortcomings, 
for some of which no doubt the personal 
equation would account. He believes it his 
duty then to record his opinion that this is 
the best book of the kind with which he is 
familiar. For system, plan, graphic pre- 
sentment and accuracy of exposition it is 
admirably adapted for school use; its 
method is that of the future. 

J. Donovan, 


THE PHOENICIANS IN ARCADIA. 


Essai de Méthode en Mythologie Grecque: 
De l’ Origine des Cultes Arcadiens: Vicor 
Bérarv. Paris: Thorin. 1894. 12 fr.50. 


No part of Greece is more interesting to 
the mythologist than Arcadia, the home of 
beliefs and customs which seem to belong 
to the earliest strata of Greek religion. 
Hence Arcadia, with its stone-worship and 
were-wolves and human sacrifice, has been 
a kind of happy hunting-ground for the 
student of folk-lore and anthropology. Fol- 
lowing quite different methods of interpre- 
tation, W. Immerwahr has recently pro- 
duced a valuable monograph on Arcadian 
religion! Immerwahr is an adherent of 
the ‘local’ school, who, working on lines 
laid down by H. D. Miiller, try to disen- 
tangle the local and tribal elements from 
the national ‘Olympian’ religion. The 
‘service which Wide, Immerwahr, Tumpel, 
and others are rendering in this direction 
cannot be too highly estimated; when 
the religious systems of every Greek 
country, city, and sanctuary have been 
examined as fully as Arcadia has been 
_ treated by Immerwahr, or Laconia by Wide, 
_ we shall have a proper arrangement of mate- 
rials for the general study of Greek myth and 
ritual, In the present learned and lucidly 
” 1 Die Kulte und Mythen Arkadiens. 1, Band. Die 
Arkadischen Kulte. Leipzig. 1891, 


. 





written volume, M. Bérard, while fully 
appreciating the work of these scholars in 
classification, declares himself dissatistied 
with the theories which they deduce from 
their materials. Immerwahr, like Roscher 
and his collaborators, works on the ‘ Aryan 
hypothesis,’ assuming that the bulk of 
Greek mythology was evolved without the 
aid of foreign influence. M. Bérard, on the 
other hand, is one of the uncompromising 
Orientalists, the most prominent of whom 
is Otto Gruppe. When the second part of 
Gruppe’s great work appears,” mythologists 
will be in a better position to examine the 
theory of wholesale and universal borrowing 
of Greek religion from the East. 

In his Hssai de Méthode M. Bérard has 
attacked the very stronghold of the Aryan 
school. The Arcadians, who lived before 
the moon, have hitherto been supposed to 
have developed their religious customs and 
beliefs with little, if any, direct foreign con- 
tamination. Their isolation in the centre of 
the Peloponnese, their pastoral non-maritime 
life, as well as the many primitive ard savage 
elements in their religion, seem strong argu- 
ments in favour of the generally accepted 
theory. Itis, indeed, admitted by historians 
that at some early period the Phoenicians 

2 Die griechischen Kulte und Mythen in thren 
Bezichungen zu den Orientalischen Religion. Bad. I. 


1887. 
¥ 2 


68 THE CLASSICAL REVIEW. 


had trading settlements on the coasts of 
Greece,! although the extent of their in- 
fluence and the date of their occupation 
cannot be estimated with any certainty. 
Probably, however, a certain amount of the 
Semitic religion survived the departure of 
the Phoenicians from Greece, and became 
incorporated with the Hellenic religion. 
Semitic influence has been pointed out in 
the mythology of Thebes (Cadmus and 
Europa, Ares and Harmonia, Heracles, &c.), 
Corinth (Aphrodite, Melicertes), Cythera 
(Aphrodite), Attica (myths of Theseus’ com- 
bat with the Minotaur and the Marathonian 
bull, the Amazons, &c.). Inthe Peloponnese, 
survivals of Oriental cults have been traced, 
with more or less probability, at Patrae, 
Sparta, Amyclae, and Olympia (see Abbott 
p. 54). But the historians seem to have 
excepted Arcadia from the sphere of 
Phoenician influence. At the most, they 
have suggested that the names of a few 
Arcadian places (e.g. Macariae, near Megalo- 
polis) have a Phoenician sound. M. Bérard, 
however, boldly claims a Semitic origin for 
nearly all the Arcadian ‘gods with the 
ritual and mythology attached to them. 
Pan and Selene alone appear to be excluded 
from his list, as primitive ‘Pelasgic’ deities.” 
In an interesting introduction he urges that 
the theory of a Phoenician settlement in the 
heart of, Arcadia presents no difliculty ; his 
arguments are partly philological and rely 
on the supposed traces of Semitic names of 
persons and places in Arcadia ;? but none 
of these names are very conclusive, and we 
may reasonably doubt the direct Phoenician 
origin of Macareus (son of Lycaon), the 
rivers Syros and Malous, and some other 
place-names for which the same claim is 
made. 

M. Bérard argues that there is no in- 
herent improbability in his assumption that 
a Phoenician trade-route passed through 
Arcadia, the foreign merchandise being 
exchanged for Arcadian cattle, timber, and 
slaves. The traders went up along the 
Eurotas and followed the course of the 
Alpheus to the sea, thus skirting Mt. 
Lycaeus. The first section of the book 
(pp. 49-93) is devoted to the principal god 
of this region—Zeus Lycaeus—whom the 
author identifies with a Phoenician Baal. 
He draws attention to the following points 
of similarity in the character of the two 


1 Mover’s Phoenizicr, pp. 47 ff., Curtius I. ch. ii., 
Duncker Hist. Greece I. ch. iii., Holm I. ch. ix., 
Abbott 1. pp. 50 ff. 

aePiO2e! 

8 Pp. 17—20. 


deities. (1) Human sacrifice is inseparable 
from the cult of Zeus Lycaeus, and may be 
compared with the like sacrifices to Semitic 
gods (e.g. to Meleart at Tyre, &e.). To this 
it may be replied that human sacrifice is 
not confined to the Semites, but is a common 
feature of savage religions. (2) No image 
of Zeus is mentioned at, any of his three 
sanctuaries, on Mt. Lycaeus, at Megalopolis, 
and at Tegea. This is an ex silentio argu- 
ment; but granting that the god was 
worshipped in the form of a stone, as 
M. Bérard suggests, it does not by any 
means follow that this stone was a Phoe- 
nician betyl. Stock and stone worship is 
so universal that it is dangerous to draw 
any ethnological conclusions from its oc- 
currence, unless, as in the case of the 
Paphian Aphrodite, worshipped under the 
form of a conical stone, the Semitic origin 
of the cult can be proved on other grounds. 
(3) Zeus Lycaeus had no vads or temple 
proper, but only a réewevos and Pwpos. 
M. Bérard gets rid of the difficulty in Thue. 
v. 16 (jutov THs oixias Tod iepod TOTE TOV Atds 
oikotvra) by the explanation that Pleistoanax 
occupied half the house belonging to the 
téwevos: ‘lautre moitié fut réservée sans 
doute pour le matériel, le personnel, ou les 
commodités du culte.’ He compares Thuc. 
i. 134 és olknua ov péya 0 Hv Tov iepod éveAOuv. 
The same criticism applies to this argu- 
ment; the absence of a temple may be a 
common feature of Semitic worship, but 
almost any page of Pausanias shows that 
an altar, without a temple, sufficed for the 
ritual of genuinely Greek gods, if the 
Orientalists will allow that any god is 
genuinely Greek. (4) There is more weight 
in the argument which M. Bérard draws 
from Pausanias’ description of the altar on 
Mt. Lycaeus: apd dé tod Bwpod Kioves dvo 
ws ert dvicyovta éotynKacw HALov, aETOL oe ér 
avtois émixpyco Ta ye ert TadaoTepa ézre- 
wotnvro.t In these two pillars he sees Jachin 
and Boas, the columns which stood before a 
Semitic temple ; while the eagles sculptured 
in relief upon them (éreroinvro, not Kanvrar) 
are really winged sun-discs. At Megalo- 
polis, the precinct of Zeus contained two 
altars, two tables, and two eagles.® Both 
at Megalopolis and on Mt. Lycaeus the 
precinct was an aBarov; the author com- 
pares the sanctity of certain mountains, 
among the Semites, as Sinai and Carmel, 
but the idea is, of course, by no means 
peculiar to the Semitic peoples. 
4 Paus, viii. 38, 7. 


° Paus. viii. 30, 2 Bwpol ré ciot Tod Ocod Kal Tpdme- 
(at d00 Kal aerol rats TparéCats toot. 





THE CLASSICAL REVIEW. 


On the whole, it must however be con- 
ceded that M. Bérard has made out a strong 
case for his Semitic theory, as far as 
Zeus Lycaeus is concerned; and if Zeus 
Laphystius (at Alos and in Boeotia) and Zeus 
Apomyius (at Olympia) are often identified 
with Phoenician Baals, there is at least as 
much probability for the same identification 
in the case of the Arcadian god. And yet 
there are serious difficulties, which should 
prevent us from too hastily accepting M. Bé- 
rard’s theory. From a historical point of 
view, it might be thought curious that no 
hint has been preserved of the foreign origin 
of the cult, as in the story of Cadmus; 
nor is there any philological evidence, such 
as has been adduced for the Oriental origin 
of Melicertes (i.e. Melcarth), Heracles (? Ar- 
chal), or Zeus Apomyius (Baal Zebub). 
For we can hardly doubt that Zeus Lycaeus 
means the Wolf-Zeus or at least Zeus of 
the Wolf-mountain, although M. Bérard and 
Immerwahr both favour the old derivation 
from ,/luc, ‘Light-god.’ But it must be 
allowed that the Wolf-god might still be 
Phoenician, as Robertson Smith (fel. of the 
Semites, p. 88, new edition, quoted by Bé- 
rard) shows that the belief in were-wolves 
was held by the Semites. 

I have examined the case of Zeus Lycaeus 
at some length, because the theory of his 
Semitic origin is at least tenable, though it 
would appear to be by no means proved. 
But the author is not content with an 
isolated Baal; he argues (p. 93) that 
Semitic deities were worshipped in trinities, 
consisting of a god, a goddess, his wife, and 
a young god, their son; e.g. at Sidon we find 
Baal Astarte and Eshmun, at Tyre, Baal 
_ Astarte and Melcarth. The origin of most 
of the Arcadian gods and goddesses is to be 
sought for in one or other of the members 
of this trinity. Again, each member of 
the Semitic trinity has a triple aspect ;1 but 
_ the Greeks formed a separate deity out of 
each separate aspect or title, so that Zeus, 
_ Dionysos, and Poseidon are really one and 
the same Baal. Hera, Apbrodite, Demeter, 
and Artemis are all forms of the great 
goddess in her celestial terrestrial and 

infernal aspects. The young god (Adonis, 
~ Thammuz, Eshmun, or Melearth) divides his 
unctions and powers between Hermes, 
Heracles, and Asclepius. The nomina have 
become numina with a vengeance. Even 
the ‘black Demeter’ of Phigaleia resolves 
herself into an Astarte. The transforma- 
tion was effected by the following steps. 
Originally Astarte rode on horseback, in her 
2 P. 174. 






















ee eee 


69 


aspect as a warrior goddess. Now Syrian, 
like Egyptian gods, often bore the head of 
their sacred animal en guise de cviffure, 
e.g. Aphrodite carries a dove on her head 
in Cyprus; M. Bérard therefore concludes 
(p. 120) that the human face of the goddess 
gradually disappeared, and the head of the 
animal descended on to the shoulders of the 
goddess. This theory seems hardly likely 
to supersede the explanation of the anthro- 
pologist, that the horse-head is the last 
relic of a goddess who was incarnate in the 
form of ahorse. We need not be surprised, 
after this, to find from the author's Intro- 
duction (pp. 44-45) that he has a very 
qualified admiration for the methods of folk- 
lorists. In fact, M. Bérard has tried to 
prove too much ; he has been carried away 
by his anxiety to find a place for each 
Arcadian deity in some aspect of a ‘triple 
triad’ of Oriental mythology. Let us con- 
sider, for example, his treatment of Le dieu 
fils. His theory, as I have already men- 
tioned, is that different characteristics of 
one Semitic god produced Hermes, Heracles 
and Asclepius. ‘ Arcadia,’ he says, ‘is the 
country of Heracles; it is the scene of a 
great number of his exploits, and the 
Arcadians are his usual companions in all 
his enterprises.’ Now it may be granted 
that Heracles became readily acclimatized 
in Arcadia; but there can be no doubt that 
he was a Theban god before he became 
Arcadian: in other words Heracles, like 
Aphrodite, may have been originally Semitic, 
but he was introduced into Arcadia after 
he had become hellenized ; his presence in 
Arcadia has nothing to do with the assumed 
Phoenician occupation of that district. 
With regard to the Arcadian Hermes and 
Asclepius, there is no authority for the 
theory of their Oriental origin. Hermes in 
Arcadia was primarily a phallic shepherd- 
god, giver of increase to the flocks and of 
luck to men; if he was borrowed from the 
Phoenicians his Semitic prototype must 
also have been a god of flocks and fertility. 
Yet we are asked to believe that ‘il repré- 
sente, dans le panthéon hellénique, une 
conception trés voisine du Verbe sémitique,’ 
p. 276. Such points of similarity as may 
exist between Hermes and Merodach (¢.g. 
both were patrons of letters) are no proof 
that the Arcadians borrowed Hermes from 
the Babylonian god. A likeness to Mero- 
dach might be traced in the Olympian or 
Pan-hellenic character of Hermes, but it 
certainly cannot be extracted from his 
Arcadian aspect. This likeness is probably 


2 Pp, 272 


70 THE CLASSICAL REVIEW. 


accidental : but granted that the conception 
of a Merodach or a Nebo may have influenced 
the Greek Hermes, the fact that his special 
Arcadian characteristics are not those of 
his assumed original proves that the Arca- 
dians did not borrow him directly from 
foreign sources. Lastly, it is true that 
Asclepius is sometimes a juvenile deity and 
is generally affiliated to Apollo; so M. Bérard 
calls him a ‘dieu fils.’ But, as Miss Har- 
rison points out,! this idea was only de- 
veloped when the cult of Asclepius came 
into conflict with that of the Dorian Apollo. 
It was necessary to find a connexion between 
the two gods of healing; so the old dream- 
oracle god gave way to Apollo, and became 
his son. Asclepius, moreover, was not 
originally Arcadian, as M. Bérard appears 
to claim,? but Thessalian. 

The conclusions which the author draws 
from his study of Arcadian religion are 
summed up on p. 323. He distinguishes 
three periods in its development :— 

(1) It appears that originally there was a 
simple Pelasgic religion in which Pan and 
Selene, the sun and moon, were the only 
gods worshipped. 

(2) A Semitic religion succeeded this 
primitive nature-worship: Zeus Lycaeus 
took the place of Pan upon his mountain. 

(3) Hellenic period. The Greeks ration- 
alized .and analysed the Oriental religion 
and so produced the anthropomorphic pan- 
theon of the historic Arcadians. 

The ‘Semitic period’ cannot be con- 
sidered as proved, or even probable, if the 
above criticisms on M. Bérard’s arguments 
are accepted. As to the ‘Pelasgic period’ 
we know as little about the religion as 
about the race of the Pelasgi. There is no 
objection to calling Pan Pelasgic if that 
term is preferred to ‘Hellenic.’ But a 
protest must be made, when not only 
M. Bérard? but Immerwahr‘ follow 
Preller ® and the old school of mythologists 
in assuming a solar origin for Pan. For if 
Mannhardt did not write in vain, he surely 
proved beyond dispute that Pan, like the 
Satyrs and Fauns of Italy, the Urisks 
of Scotland, and the Ljeschie of Russia, 
was a wood spirit conceived of in the form 
of a goat.® But, as there seems to be a 
perpetual recrudescence of the solar theory 
i connexion with Pan, in spite of Mann- 
hardt and Mr. Frazer, it may not be useless 
: Myth. and Mon. of Ancient Athens, p. 324, 


2 Pyi270 

3 Pp. 61—62 &e 

4 op. cit. p. 204. 

° Preller-Robert [.2 pp. 738 ff. 

8 Antike Wald- und Feldkulte, ch. 8. 


to examine the arguments which Immerwahr 
and his supporters bring forward. Pan has 
his flocks in Arcadia; Helios and Apollo 
have their herds ; therefore Pan is a sun- 
god. But Pan, as the god of a shepherd 
people, naturally protects the flocks of his 
worshippers. The herds of Helios or 
Apollo are no parallel, whether they belong 
specially to the Sun-god, or whether (as in 
Hom. hymn. Mere. 71, Oedv paxdpwv Bodes) 


they are the common property of the gods, 


and are merely tended by Apollo. Again 
Pan is the son of Ether, according to one 
tradition (Mnaseas); but this does not 
appear to be an early genealogy ; and indeed 
the parentage of the god is too indis- 
criminate to draw any conclusions there- 
from. His father is Uranus, Zeus, Hermes, 
or Apollo; his mother Callisto, Oenoe, 
Penelope or the daughter of Dryops, ac- 
cording to various accounts (see Preller- 
Robert, where an even longer list is given). 
His epithet aidAos in Macrob. Sat. i. 23 may 
refer to the sun in the mind of Macrobius 
himself, but is certainly no proof of an 
original solar character. Tmmerwahr points 
out that Helios and Pan were worshipped 
together at Sicyon ; but unless the men of 
Sicyon, like Pentheus, saw two suns, this 
fact might as easily show that Pan was 
not the sun. It may be true that Helios 
and Apollo were little worshipped in Ar- 
cadia,’ but this is purely negative reasoning 
and does not concern Pan ; nor isa developed 
sun-worship by any means universal. The 
solar theorists make much of the fact that 
Pan and Selene were worshipped together 
in Arcadian caves* and that Pan won the 
love of Selene by an artifice.? But both 
Pan and Selene! inhabit caves, so that 
there is nothing wonderful that the two 
deities should have a cave incommon, I 
find that this obvious explanation is also 
given by Roscher (Selene, p. 151), who 
suggests an additional link between Pan 
and Selene;.in his view, Pan is merely a 
divine counterpart of the Arcadian shepherd- 
hunter, and is connected with the moon 
because a moon-lit night is favourable for 
watching the flocks and for nocturnal. 
hunting expeditions." 


7 Immerwahr, p. 205; Berard, p. 62. 

8 Porphyr. de Antr. 20. 

® Macrob. Sat. v. 22, niveis velleribus se circwm- 
dedit. Of. Virg. Georg. ili. 391. 

1” Selene visited Kndymion, the shepherd and 
hunter, ina cave on Latmos. The absence of the 
moon was accounted for by the primitive explanation 
that the moon-goddess was hidden in her cave, 
Preller-Robert I.* p. 445. 

1 op. cit. pp. 162 ff. 





THE CLASSICAL REVIRPW, 7) 


If Pan, therefore, as seems certain, had 
no connexion with the sun, there is little or 
no evidence to support the theory that the 
‘Pelasgic’ religion was confined to a simple 
worship of the heavenly bodies. M. Bérard’s 


reconstruction of Pelasgic beliefs is as du 
bious as his theory of a * Phoenician period,’ 
influencing the whole of Arcadian ritual 


and mythology. 
EK. E, Sixes. 


RECENT EDITIONS OF HYPERIDES. 


Hyperides, the Ovations against Athenogenes 
and Philippides, edited with a Translation 
by F. G. Kenyon. London, George Bell 
and Sons, 1893. 5s. net. 

Hyperidis Orationes Sez cum ceterarum 
fragmentis edidit F. Buass; ed. tertia, 
insigniter aucta. Leipzig, Teubner, 1894. 
2m. 10pf. 


Mr. Kenyon has earned the gratitude of 
many scholars at home and abroad by the 
skill with which he has deciphered, and 
the promptitude with which he has pub- 
lished, the important Greek papyri which 
have recently been secured by the British 
Museum. The object of his present volume 
is to ‘make available for readers, in an 
accessible form, the two most recently 
recovered orations of Hyperides.’ Of these, 
the oration against Athenogenes has been 
published by M. Revillout and others, while 
the fragment of that against Philippides 
was first edited by Mr. Kenyon in a volume,! 
which he modestly describes as ‘ containing 
a large quantity of other matter, which a 
reader may or may not desire to possess.’ 
Mr. Kenyon now supplies us with an 
interesting Introduction, a fairly satisfactory 
Text, and an eminently readable Translation, 
while the general attractiveness of the book 
is further enhanced by a Facsimile of nine- 
teen lines of the Speech against Atheno- 
genes from the papyrus in the Louvre. This 
MS is not later than the end of the second 
century B.C. ; it is thus the oldest extant MS 
of any classical Greek work yet discovered, 
with the exception of the fragments of the 
Antiope and the Phaedo. 

The recovery of the Speech against 
Athenogenes is particularly welcome as the 
author of the treatise On the Sublime couples 
it with the defence of Phryne as an example 
of a style in which Hyperides was superior 
even to Demosthenes. Athenogenes is an 
Egyptian resident in Athens, who has a 
slave named Midas (probably a Phrygian). 


1 Classical Texts from Papyri in the British 
Musewm, 1891, 


Midas, who has two sons, is manager of a 
perfumery belonging to his master, Hy- 
perides’ client wishes to acquire possession 
of the younger son, and is informed by the 
slave-boy’s master that, if he wants to buy 
the boy, he must buy his brother and father 
as well. The original proposal to pay for 
their freedom only is cunningly changed by 
their master into one for buying them right 
out. When the transaction is completed, 
the purchaser finds himself saddled with 
heavy liabilities incurred by Midas, the 
full extent of which he now learns for the 
first time. He accordingly brings against 
Athenogenes an action which has, with 
great probability, been identified as a dicy 
BraBys. The intermediary in negociating 
the bargain, in its original form, is a person 
of questionable character named Antigona, 
whose success in deluding the plaintiff is 
complete. The plaintiff's own character is 
obviously not high; and, having formally 
consented to the purchase and actually paid 
the money, he has in point of law a weak 
case. There was all the more reason why, 
in a matter requiring skilful and delicate 
handling, he should seek the aid of an 
expert like Hyperides. 

The Speech against Philippides is con- 
cerned with a ypad) tapavéuwv. Philippides 
had moved a vote of thanks toa certain body 
of zpdedpor for the manner in which they had 
discharged their duties as the presiding 
committee of the éxxAnoia. The mpdedpor 
had put to the vote a proposal in honour of 
Philip. The proposal was irregular, but it 
had been put and carried under pressure. 
To screen the zpdedpo from the consequences 
of this irregularity, Philippides, a member 
of the Macedonian party, proposed to vote 
a crown to the zpdepo ‘for their upright 
and legal action.’ Hyperides attacks this 
proposal as illegal. A point of interest may 
be found in the fact that among the friends 
of Philippides is one Democrates of Aphidna, 
who belongs to the same deme as Harmodius, 
and is a descendant either of Harmodius 
or (less probably) of Aristogeiton. In a 


72 THE CLASSICAL REVIEW. 


note on p. 51, Mr. Kenyon, by the way, 
describes Democrates as ‘a descendant of 
either Harmodius or Aristogeiton, probably 
the latter, who appears to have belonged to 
the same tribe of Aphidna.’ But it was 
Harmodius, and not Aristogeiton, who 
belonged to Aphidna. The note is easily 
corrected by altering latter into former ; and 
tribe into deme. ‘The speech against Philip- 
pides adds to our knowledge of the privi- 
leges enjoyed by the descendants of the 
‘tyrannicides’ by informing us of a law 
‘forbidding any one either to speak evil of 
Harmodius and Aristogeiton, or to sing 
insulting songs about them’ (doau ért ra 
kakiova). 

In the text of the Athenogenes, col. i 14, 
the wily Antigona is described as qgevaki- 
Lovoor i. 22%. | a rtatra. As the proposed 
insertion of dravra involves a hiatus, and 
neither this nor zédvra is sufficient to fill 
the gap, Mr. Kenyon supplies [ra pdracja 
tavra. If an adjective is needed at all, I 
may suggest, as an alternative, [raraywy]a 
taira, which contains exactly the same 
number of letters as Mr. Kenyon’s con- 
jecture. ézaywyés is particularly appro- 
priate to the seductive blandishments of a 
person of Antigona’s class. Twice in Lu- 
cian’s Dialogi Meretricii (1, 2; 6, 3) a 
similar character éraywydv pedi (cf. Hat. 
3, 53, Ta éraywydtata éyew, Thue. 6, 8, 
eraywya Kat odk ddn67, Dem. Neaer. 70, éxa- 
yoyous Adyous).1 In col. viii 24 we have 
next to nothing in the text answering to 
the rendering : ‘plucking me like a bird 
taken in a snare’; it is not until we turn 
to the critical note that we find the corre- 
sponding Greek :—déo7ep bro[ xe(piov év rodoc- 
TpaBy KarleAnupevov, which might well 

ave been printed in the text. In col. ix 
14 [oid byes cilpeiv is clearly less good 
than [ovdev dyes élpciv. obdty yids déyenv, 
and the like, occur nine times in Demo- 
sthenes (18 §$ 23; 27 § 26; 29 § 5; 40 
S$ 21, 53; 48 § 51; 58 §§ 12, 36; 59 
§ 125); otde tyes eipioxew never. In col. 
x 17 the editor accepts :—[édév ru éy]abov 
T™paen 1 épylaci av eip[olol[dcav eéyy, told 
KEKTHLEVOV abTOV yey ver[au], with the trans- 
lation: ‘if a slave effect a good stroke of 
business or establish a flourishing industry, 
it is his master who reaps the profit of it.’ 


* But for considerations of space, I should have 
preferred to propose gevariCovca [xatamwaTrar|a or 
[kamrataolaraidra. pevariCew and etamaray are coupled 
in Dem. 19 § 29; 21 § 204 ; 23 § 195. damdrn occurs 
in § 27 of the same speech of H yperides, and amarav 
in fragm. 21 ; but here, as elsewhere, etamaray is 
more common (i 6, 12 ; iii 86; iv 5). Both verbs 
are found with cogn. acc. 


eipoovcav is supported by Revillout, Diels 
and Weil; but the authority for such a 
word in the Attic Orators is ni/. Plato has 
evpova Of ‘successful progress,’ and it is also 
to be found in Alcidamas, wept Sogucray, 17, 
where it means the same as eizropia, and has 
possibly been substituted for it, edzopia and 
the like being in constant: use in this de- 
clamation (cf. $ 3 etmopia, $$ 19, 24, 34 
evropos, S$ 6, 13 eizdpws, $ 26 edropnpa, 
§$ 17 dropos and §§ 8, 15, 21 bis dzopia). 
Polybius has evpoia toy rpaypdtwv, and tov 
Tpaypatwv evpoovvtwy (quoted in L. and 8.) ; 
but the Orators have nothing of the kind. 
We should therefore prefer the proposal of 
Blass :—evp[y] 6 ofixérns]. In col. xvii 6 
Mr. Kenyon prints his excellent proposal 
[rodrov broxetprov| ciAndores, Which may be 
supported by Lysias 4 $ 5, and Dem. 23 
§ 175; but, if (as we learn from Blass) 
there is only. space fer ‘about thirteen 
letters, we are reluctantly compelled to 
acquiesce in the less interesting suggestion 
of the German editor :—[rotrov tyels viv] 
ciAndores. ; 

In the Speech against Philippides, col. i 
19, [ev]6, which can only mean ‘there’ or 
‘then’ (or ‘where’ or ‘ when’), is unsatis- 
factorily proposed in the sense of ‘here,’ as 
though it were synonymous with évravéa or 
evOdde. The passage is intricate, and a 
perfectly satisfactory restoration far from 
easy ; but this, at any rate, cannot be right. 
In the next sentence, however, Mr. Kenyon 
has since shown his skill by suggesting the 
reading now adopted by Blass :—eis éorépalv 
delirv[ylowy (instead of [ovlvzA[do|owv) as 
tulas epxlera. (The subject is Democrates, 
who, as a descendant of Harmodius, is 
entitled to dine in the Prytaneum.) In col. 
v 112, & pe capa aOdvarov six[etAn|pas 
éocofar is translated, ‘you were foolish 
enough to suppose that a single individual’s 
life would last for ever.’ The ‘life’ is the 
life of Philip, and Mr. Kenyon (in his Intro- 
duction) rightly holds that Philip is still 
alive, while Kohler supposes the speech was 
delivered after his death. Consistently with 
the former view, it would perhaps have been 
safer to translate the verb not as an aorist, 
but as a perfect, best represented in English 
by a present :—‘ putas (minime putasti vel 
putabas),’ as observed by Blass on p. lili of 
his edition. In col. viii 188 Mr. Kenyon has 
been prompted by Blass to print rév pevdo- 
paptupiwv (of the second declension), instead 
of Wevdopaprupidv (of the first). Blass refers 
to Pl. Theaet. 148 B, evoyos rots Wevdopnap- 
tupios. To this one may add Aristotle, 
"AG. wor. 59 § 6, Ta Wevdopapripia <Ta> €k 








THE CLASSICAL REVIEW. 73 


"Apetov rayov, and Cratinus quoted by Pollux, 
viii 31 :—wWevdopaprupia’ Kparivos dé kal Wevdo- 
papripiov eipyxev. Pollux clearly regards 
the form in -ta as the normal form, though 
it is never actually found in the singular 
except in the earlier texts of Isaeus 12 § 6 
and Dem. 41 § 16, where the acc. sing. -fav 
is now altered into the gen. pl. -ay. The 
latter form is printed by Scheibe in at least 
ten passages of Isaeus, and by Blass in no 
less than thirty passages of Demosthenes, 
while in Dem. 57 § 53 we have, as clear 
evidence for the first-declension form, év 
Wevdopaptupias. The form in -iay is also 
recognized in Bekker’s Anecdota, p. 194, 27. 
Thus we have only three certain instances 
of the neuter form, against forty instances 
of the feminine, unless, indeed, we are pre- 
pared to alter all of these into the neuter. 
The fact is that the forms are alternative ; 
but the feminine form is much more common 
than the other. Just so, paprivpiov exists by 
the side of paprupia, though with a slight 
difference in usage. 


With the exception of the Speech against 
Athenogenes, now in the Louvre, all the 
MSS of Hyperides have found their way 
to the British Museum ; and the texts of 
all have now been united for the first time 
in a single volume by Professor Brass. In 
the language of the ancient epigram, we 
may now say that all the papyri of the most 
brilliant of the Attic Orators, ozopddes zoxa, 
viv dpa maoat | évti pias pavdpas, évTl pds 
ayédas. Professor Blass is to be congratu- 
lated on the publication of the third edition 
of his work. The first appeared in 1869 ; 
the second was an improvement on the first, 
and the third shows a further advance in 
many points of detail, besides containing 
both the newly-discovered speeches. The 
new material includes the ‘ Tancock frag- 
ments’ of the speeches against Demosthenes 
and for Lycophron, published by Mr. Kenyon 
in the Classical Review, vi 288, and the 
‘Raphael fragments’ of the former. One 
of these last (pp. 11—12 of Blass) supplies 
us with a parallel to Aeschines and Plutarch. 
Aeschines, 3, 209, says of Demosthenes, 
exhirov pev TO dot, ovK oikeis ws Soxets ev 
Tletpacet, GAN’ eéoppeis ex tis TéAews. Hy- 
perides (as restored by Mr. Kenyon) borrows 
this phrase, and says :—ov« oikeis [ év He |tpace?, 
GA(A)’ eSoppets ek tis woAews. Again, Plu- 
tarch, Comp. Dem. et Cic. 3, describes Demo- 
sthenes as having sums invested in loans on 
bottomry, daveiLovros éri vavtixois. Hyperides, 
addressing Demosthenes, says [viv d@ vav}re- 
kois épydfy. In col, xxiv of the same 


speech (p. 16) Blass now reads. of dé vol por 
Tots} pev ddixood| ow amh]a, rois be bw! podoxod 
oly dexarrAa [ra OpA)jpara t[p joorar| rov- 
aw] drodidvar. It is interesting to note 
that the substitution of drA@ for burda (the 
reading of ed. 2) is due to the new light 
derived from Aristotle, ’A@. rod. 54 § 2, dy 
d€ ddicely katayvoow, ddixlov tysdaow, drori- 
verat 5€ todro dAotv. Again, in col. viii 
3—4, ahi (7) yypws] eve[Kev] } vooov } pave 
gives us one more reason in favour of insert- 
ing évexev, or évexa, after éiy pi pavdy 
yipws.in’AG. rod. 35 § 2, instead of retaining 
paviov 7 ynpov and regarding them as rare 
and exceptional examples of participles. 

In pro Euxenippo, § 19, as well as in the 
Funeral Oration, § 27, and in fragment 
219a, we find the word é¢dduv. This enables 
us to correct the statement in Liddell and 
Scott, that this word is rarely found in the 
singular. It may be added that, in Demo- 
sthenes, while the plural is used seven times, 
the singular is also used in as many as five 
passages (19 § 158, 25 § 56, 34 § 35, 
53 §§ 7, 8). 

In the Philippides, p. 536, the lacuna in 
éy® 6é.....v may perhaps be filled up by 
reading éy& dé [rovva]y[réov]. On p. 56 the 
proposal kali yxopo]y ioras yeAwroz[ ody) is 
confirmed by p. 58, xopdaxilwv kal yeAwro- 
TOLOV. 

In the Funeral Oration, p. 78, Biicheler’s 
suggestion padpt|vs axprBijs 6 x|pdvos may be 
supported by Lycophr. § 14, 6 rapeAnAvbas 
xpoves pdptus éoriv—dxpiBéoraros. Other- 
wise, one might be inclined to propose 
papt[us txavds 6 x|povos. txavds is an epithet 
of paprupia in Plat. Symp. 179 B, and of 
texunprov in Gorg. 457 D and Phaedo 70 D, 
and is joined with rexunpiocac in Thue. i 9 
§ 3. But Hyperides himself has ovre 6 
xpovos txavds, only twenty lines below this 
passage, and this may weigh against my 
suggestion, In col. iv wit. I still adhere to 
a proposal made in the course of a review 
of the first ed. (Academy, 1870, p. 221), 
rerarevopevyvy Kal [de xatlerry[yvliay. No 
other word meets the case as well as dée, 
which is found in the dat. in Dem. 4, 45 and 
21, 124. In col. x 7-10 one is glad to see 
the manuscript reading iva. . wvrag now 
represented by Sauppe and Kayser's als 
ai[ve|ov raéw instead of Cobet’s «ls dpe 
raéw. The fact, which I had occasion to 
point out in the above review, that the o 
after the lacuna is really altered into 0, 18 
accepted by Blass as decisive. 

The volume closes with an excellent index 
prepared by one of Professor Blass’ pupils, 
H. Reinhold, Among the items in this 


74 THE CLASSICAL REVIEW. 


index which ought to be noticed in future 
editions of Liddell and Scott are, évoedev 
rive eis viv (Vv 26), ‘to entrap iuto a sale’ ; 
katatéuvev twa (v 12), ‘to cry a person 
down’; zaidaywy<iv (v 2), ‘to delude’ ; 
and zpoomepikorrew (Vv 2), ‘to appropriate 
in addition,’ or (if an accusative of the 
person follows) ‘to plunder afresh.’ This 
last is a new compound. 

Englishmen have done much for the 


recovery and restoration of the text of 
Hyperides ; and scholars in France, Ger- 
many, Holland, Italy and Sweden have 
contributed a good deal towards the study 
of the subject. Both of the works here 
noticed will doubtless serve to extend the 
interest which it has already inspired in 
England and elsewhere. .. 
J. E, Sanpys, 


BELLING ON TIBULLUS. 


Kritische Prolegomena xu Tibull. Hz. 
BELLING. Berlin, Weidmann: 1893. 
8vo, pp. 97. 3 Mk. 


Quaestiones Tibullianae, scripsit HEnRicus 
Bexuinc. Beilage zum Jahresbericht des 
Askenischen Gymnasiums. Berlin : 
1894, Progr. no. 51, 4to. pp. 26. 


THESE two pamphlets are a remarkable 
performance. ‘They contribute new ideas 
towards the solution of the most vexed 
questions of Tibulline criticism. The most 
important of these may be thus distinguished 
and arranged. ‘Except the lost Cujacian 
fragment (F) all the known codices of 
Tibullus descend from a single copy of an 
injured exemplar (t); the injuries of this 
exemplar were chiefly in the first and last 
lines of a page; the copyist of t supplied 
the missing portions where and as he could ; 
not only did he interpolate here, but also 
where the exemplar contained repetitions 
or redundancies that offended him.’ The 
following statement from pp. 42 sg. of the 
first pamphlet will show how the author 
applies the first three theses to the explana- 
tion of the existing text :— 

“ After i. 2, 25a pentameter was lost in t. 
The scribe went on with the next hexameter 
25a. In i. 4, 44 only the end was left 
‘imbrifer arcus aquam.’ The scribe supplied 
‘uenturam admittat [or annuntiat].’ In i. 5, 
33, only the end was left ‘hunc sedula 
curet.’ The scribe supplied ‘et tantum uene- 
rata uirum. Ini. 5, 47 only the beginning 
was left ‘haec nocuere mihi.’ The scribe 
supplied ‘quod adest‘huic diues amator.’ In 
i. 6, 42 only the end was left ‘stet procul 
ante ula.’ Zhe scribe supplied ‘stet procul 
aut alia. Ini. 6, 72 only the end was left 
‘proripiarque uias.’ Zhe scribe supplied 
‘immerito propriis. In i. 7,56 only the 
beginning was left ‘augeat.’ The scribe 


supplied ‘et circa stet weneranda senem.’ 


After i. 10, 25 a whole couplet was lost. The 
scribe went on with the pentameter 26. In 
ii. 1, 58 only the beginning was left ‘dux 
pecoris.’ Zhe scribe supplied ‘hircus auxerat 
hircus oues.’ Inii. 2, 21 only the end was left 
‘prolemque ministret.’ Zhe scribe supplied 
‘hic ueniat natalis awis.’ After ii, 3, 14a 
apparently a whole pentameter was lost. 
The scribe went on with the hexameter ii. 3, 
146. Inii. 3, 14¢ apparently only the end 
was left ‘obriguisse liquor.’ The scribe 
supplied ‘lacteus et mixtus.’ In i. 3, 34 
only the beginning was left ‘imperat.’ The 
scribe supplied ‘ut nostra sint tua castra 
domo. After ii. 3, 74 a hexameter was 
lost. Zhe scribe went on with ii. 3, 76. In 
ii, 4, 22 apparently only the end was left 
‘et Coa puellis.’ Zhe scribe supplied ‘hic 
dat auaritiae causas.’ In distich ii. 4, 37 
sq. only the beginning of the hexameter was 
left ‘hine fletus rixaeque sonant.’ The 
scribe supplied ‘haec denique causa fecit ut 
infamis hic deus esset amor. After iii. 4, 
64 apparently a hexameter was lost. The 
scribe went on with the pentameter 66.” 

That in the majority of these passages 
(to which others are afterwards added) the 
tradition is corrupt will be admitted by every- 
body.! The author would further maintain 
that it has been corrupted in a particular 
way. In five out of the above passages i. 4, 
4; 5, 33; ii. 1, 58; 3, 14c; 4, 38 are metrical 
faults of the same kind; and in other 
respects they evince the same handiwork, 
e.g. in the use of the pronoun hic i. 5, 47; 
ii. 2,21; 4, 22, 37, 38 and cf. iii. 6, 23 
‘deus hic.’ It is necessary in fairness to 
the writer’s case to cite together the exam- 
ples of apparently uniform interpolation ; 

1 Except K. P. Schulze whose ignorance of 


prosody is a ground of just astonishment to the 
author, p. 11 n. 2, p, 14 n. 3 








THE CLASSICAL REVIEW, 75 


but conviction one way or another can only 
be reached by an examination of the 
passages in detail. Of these the first 
treated is i. 4, 44. The writer begins an 
able discussion by proving that in ‘picta 
ferrugine’ (‘picea f.’ wulg.) 43 and in 
‘imbrifer arcus ’ 44 the tradition is perfectly 
sound. Passing on to 44 ‘admittat’ wulg. 
is shown to be intolerable: ‘amiciat’ 
A(mbrosianus) and ‘anutiat’ V(aticanus) 
point to an ‘infitiat’ in the archetype : 
which, corrupted in other copies (for 
Belling accepts the views of Baehrens 
and Magnus that certainly more than 
one were made), has produced (through 
‘ammittat’) the ‘admittat’ of y (Lach- 
mann’s A, Heinsius’ ‘codex Archiepiscopi 
Eboracensis’) d (Lach.) e (Lach.) and 
again the ‘annuntiet’ of > (the ‘inferior’ 
MSS.). Now ‘uenturam annuntiat’ has 
two marks of an interpolation, fair sense 
and faulty diction. It is the offspring 
then not of error, but design. This, 
we may add, is confirmed by observing 
that ‘annuntiat,’ is ecclesiastical Latin with 
vulgar pronunciation (cf. Fr. annoncer). 
‘uenturam’ must fall with ‘annuntiat’ ; 
for, passing over the question of its appro- 
priateness which Belling denies, a verb is 
wanted, and a verb cannot stand in the 
position of ‘annuntiat.’ Here then it 
appears to me to be reasonably certain that 
half a pentameter has been interpolated by 
some incompetent hand. I think the same 
may be said of i. 6, 72 on which it is well 
pointed out that the interpolation is based on 
the corruptions ‘possum’ 70, [‘ putat’ and] 
‘ducor’ 71, and ‘ proripior,’ already present 
in the archetype, and of ii. 3, 14c. It 
remains to consider the grounds of these 
interpolations. First it should be noticed 
that they consist not of single words, but 
of halves of verses. The first change any 
copyist might have attempted under any 
circumstances; not so the latter. And, 
taking account of the quality of the inter- 
polator’s work, I agree with the author 
that there was some patent ground of 
offence in the text to drive him into 
original composition. Illegibility is such 
a cause, obvious and adequate, though 
others might be thought of such as 
the dislike of tautology which will be 
considered below. In these three passages 
then interpolation replacing  illegibility 
appears to be a reasonable explanation of 
our text. In the following corrupt passages 
the same hypothesis is not without some 
plausibility, i. 3, 50 where repente is incon- 
sistent with the ‘mune Toue sub domino 


caedes et uulnera semper’ though it is 
hardly credible that the interpolator would 
have hit upon that word, and it is more 
likely that he wrote ‘ reperte ' te. ‘ repertae’ 
which some of the s indicate. B. 


roposes 
‘nune mare nunc longae (or ‘ ph ) fata 
ulae properant’—a verse which has naturally 
found no acceptance. If his meaning is 
right, ‘ terrae mille pericla uiae’ would give 
it better. Ini. 5, 33 ‘uirim hunec’ shows 
something is wrong ; and Belling here and 
at 7, 56 attacks the use of ‘ ueneror,’ but, 
like other critics, I cannot see why. No 
doubt the word had a special religious 
and ceremonial application as he shows 
from a large collection of passages; but it 
was not always so limited. It would be 
curious to see how he would show that a 
word inappropriate to Messalla here is 
appropriate to ‘concubitus’ ef g. s. in Ovid 
Ars ii. 307 sg. In 5, 47 ‘quod adest huic 
diues amator’ is certainly corrupt and most 
probably interpolated ; but the mischief does 
not stop here. A couplet at least has been 
lost in which Tibullus passed from the fatal 
charms of his mistress (‘haee nocuere 
mihi’) to the deadly injury which the ‘ lena ’ 
had done him. Is it credible that in the 
same couplet he wrote ‘haec nocuere mihi ; 
<sed iam dominam tenet alter> (B., or any 
other stopgap you please) uenit in exitium 
callida lena meum’? On i. 6, 42 B.'s 
objections to previous suggestions are 
pertinent ; and his own view, and supple- 
ment ‘luminibus parcens,’ may possibly be 
right. But the hypothesis of dittography is 
not untenable ; and the second ‘ stet procul’ 
arouses suspicions. We might guess the 
original to have been ‘ stet procul aut alio 
segreget ante uiam’ or something of similar 
sense. 1. 7,56 is not above suspicion nor 
the remedy of > and Baehrens, ‘ uenerata,’ 
a certain one; but our embarrassments do 
not end here, as we have to face the uncer- 
tainties of reading and connexion in the 
following verses. Perhaps here too a 
couplet has been lost. B.’s observation that 
the pentameter would naturally refer to the 
death-bed of Messalla has been refuted by 
others ; and the note (p. 35) that cireum 
(circa) is elsewhere a ‘post-position’ in 
Tibullus is seen to be irrelevant when the 
passages are examined. Nor is our way 
clear in ii. 3, 34, though the sense of B.'s 
alternative for the MS. reading which is 
most obscure ‘exemplo uiuere disce dei,’ 
would fit the passage very well. In the 
other passages the assumption of interpola- 
tion appears to be mistaken. ii. 1, 58 I 
have discussed in the Journal of Philology 


76 THE CLASSICAL REVIEW. 


xx. pp. 313 sg. The first ‘hircus’ seems to 
be a gloss, and the line should be read ‘dux 
pecoris curtas (Waardenburgh) auxerat 
hirtus opes’ (W.). In ii. 2, 21 ‘hic’ should 
be ‘ hine’ whose sense I fear it could not 
have here ; the couplet is quite intelligible if 
we read ‘prolesque’ and in 22 ‘ut’ (>) for 
‘et’ with Baehrens. The unusual construc- 
tion ‘ ministret—ut’ is not likely to have 
been interpolated, whereas the changes ‘ et’ 
and ‘prolem’ are quite intelligible. In ii. 
4, 29 ‘addit’ or ‘indit’ should be read for 
‘hic’ (‘hine’ ¢) ‘dat,’ the Paris excerpta 
have ‘ praebet auaritiae causas’ and the last 
two words are certainly genuine. ii. 4, 38 
B.’s_ arbitrary objection to the words 
attached is that ‘Amor eis tantum rebus 
infamis factus dici potest quae ad Amorem 
auctorem’ [the italics are mine] ‘referun- 
tur.’ Why should not Love be discredited 
through no fault of his own? For ‘hic’ 
read ‘nunc’ with Broukhusius and all 
difficulty disappears. ‘hic deus’ may be, 
as Belling suggests, a reminiscence of ii. 
1, 79. The four omissions of whole lines 
can at most be assigned only corroborative 
value. They may be due to injury, but 
they certainly do not prove it. 

To sum up this portion of the argument, 
interpolation provoked by injury to the 
archetype has not been made probable for 
more than a.few passages: but this is not a 
refutation of the hypothesis of injury ; for 
the corruptions as the omissions, which are 
undoubted, may be due to such a cause. 

The theory of pagination which it is 
sought to connect with these and other 
phenomena of our MSS. is that the arche- 
type consisted of 71 leaves, half of the last 
being blank, of writing with 14 (or 13 or 
15) lines on a page. On these variations 
from it criticism has naturally fastened. 
In his Quaestiones p. 6 the author does 
indeed cite an inconstancy from the Am- 
brosianus, which twice has 21 instead of 22 
lines to the page. But the probability of 
such a fluctuation, which decreases as the 
number of lines is diminished, cannot be 
great for a page of 14. He claims—and 
herein he has had somewhat less than 
justice from his countrymen—to have the 
probability of his theory judged in the 
gross. To this end I have examined the 
MS. tradition of Tibullus with the following 
result, Leaving out of the reckoning 
merely clerical mistakes I found as a rough 
estimate that 33 cases of serious and 12 of 
less grave corruption which might be 
ascribed without improbability to injury 
in the archetype fell within the lines which 


would begin and end the page on the 
author’s theory, while 20 and 7 respectively 
were included in the intervening spaces, 
If however these corruptions were evenly 
distributed, the latter should exceed the 
former in the ratio of 6 to 1, or in other 
words the totals should be for the former 
10 instead of 45 and for the, latter 62 in 
place of 27. It may be said that a minute 
investigation, which cannot be undertaken 
here, would reveal the hollowness of these 
figures; but until then we cannot reject 
the hypothesis, however artificial we may 
think it. My own. impression is that it 
contains a vera causa, though not yet 
detached from irrelevant surroundings. It 
would certainly gain in probability with the 
acceptance of the last of the four positions 
distinguished at tne beginning of this 
article. The removal of apparent tauto- 
logy is admissible as an explanation of 
the change in ili. 6, 23 ‘quales his poenas 
qualis quantusque minetur’ (so F, ‘deus 
hic’ the rest) and iv. 1, 40 ‘hie aut hic 
tibi’ (F, ‘hic aut tibi’ the rest); 7b. 205 
‘seu matura dies celerem properat mihi 
mortem’ (so F, ‘fato’ AV which however 
introduces, though less obviously, a tauto- 
logy). B. also assumes it to explain the 
variants in the following places where it is 
improbable, arbitrary or inadmissible; i. 
1, 25 ‘iam modo iam’ altered to ‘i. m. 
non’; 43 ‘satis est’ omitted by V because 
there were dots under it in the archetype ; 
7, 49 ‘centum’ written for the first 
‘genium’; i. 3, 4 ‘modo nigra’ written for 
‘precor atra’ (but why was the first ‘atra’ 
altered and that violently instead of the 
second ?) ; 9, 40 the second ‘sit’ was altered 
to ‘sed’ (!); ii. 1, 67 ‘interque greges’ is 
the original and ‘quoque inter agros’ A 
(‘quoque inter greges’ V) descend from the 
conjecture of the scribe who objected to 
‘ereges’. before ‘armenta’ (he does not 
seem to have minded the thrice repeated 
‘inter’!); iii. 2, 24 for ‘diues’ the scribe 
suggested ‘pinguis’ which occurs in some 
s; 4, 4 ‘uotis’ was written for ‘uanis’ 
because of ‘uani’ in 3; 4, 65 omitted 
because of ‘saeuus amor’ in 66 (the omission 
is clearly accidental) ; 66 ‘saeua’ F altered to 
‘posse’ because of ‘saeuus’ (!) ; 6, 21 ‘non 
uenit’ for ‘conuenit’ and ‘seuerus’ for 
‘-os’ due to a misunderstanding of ‘nimium 
nimiumque’; iv. 1, 112a@ is omitted in some 
> because the genuine (!) double ‘saecula 
famae’ was athetized by the scribe; 5, 16 
for the same reason ‘hac’ after ‘ post hac’ 
disappeared in > and was replaced by ‘nos.’ 
I am afraid this collection only shows the 








THE CLASSICAL REVIEW. 77 


inexperience of the author in dealing with 
manuscript evidence. The rest of the 
prolegomena calls for but brief notice. In 
pp. 74—90 passages where other sources of 
corruption are said to have been at work 
are passed uuder review. Heyne’s conjec- 
ture in i. 6,7 is rightly approved and the 
appropriateness of ‘ patera’ in iii. 6, 3 justly 
disputed. In i. 1, 41 ‘igne’ is rightly 
defended ; perhaps also ‘magna’ in v. 2 of 
the same poem, though the authority for 
the easier ‘multa’ is strong. In i. 2, 19 
the question between ‘derepere’ Fris. exc. 
and ‘ decedere’ AV is one not of taste but 
authority. B.’s treatment of i. 5, 69 sqq. 
where he would have it that ‘mea fata (the 
almost universally accepted correction of 
‘furta’ MSS.) ‘timeto’ must be understood 
in the same sense as ‘metuit qui fata puellae’ 
iv. 4, 11 is entirely mistaken; 71—74 
belong to the next poem. The same may 
be said for i. 9, 23 sg. where apparently we 
should read ‘ celari’ for celanti’ and iii. 5, 
11 ‘aegros’ (uulg. ‘ignes’ bringing sick 
men to temples to steal the ‘ sacra,’ which is 
all the words could mean, would be a 
farfetched sort of villainy). This portion 
concludes with the proposition that ‘An 
mehreren Stellen haben Herausgeber das 
handschriftlich iiberlieferte ipse ohne zwin- 
genden Grund aufgegeben.’ These passages 
are 1. 2, 58, ii. 3, 59, 4, 36, iv. 13, 8, and i. 
5,74. The reader may judge for himself. 
The last five pages concern themselves with 
the composition and authorship of the last 
two (three) books. He believes that the 
‘Sulpicia Cyclus’ iv. 2—6 was written by 
Tibullus himself and that these and all the 
rest of books iii and iv were found amongst 
the effects (Nachlass) of Tibullus at his 
death. We are to suppose then that a copy 
of a panegyric of Messalla, the works of the 
unknown poetaster Lygdamus, love epistles 
which the poet wrote for Cerinthus and the 
billets doux of Sulpicia were given to the 
world as Tibullus’ immediately after his 
death and during the lifetime of his patron 
Messalla! It is a wonder we are not told 
they were edited by Messalla himself in 
one of the lapses of memory which it is 
said afflicted his later life. Like others, L 
cannot bring myself to believe the Sulpicia 
Cyclus is by Tibullus. Not only is the 
metrical treatment different (thus hyperdi- 
syllables at the end of the pentameter are ten 
times as rare as in the second book) ; but they 
impress one as the work of a different mind. 
But then I cannot believe that iv. 7 is 
by the same author, though Belling after 
Ehwald says that the construction of v. 4 


is ‘decisive.’ This is a good instance of the 
inattention with which arguments are manu- 
factured. It is quite possible of course that 
the writer of ‘ Exorata meis illum Cytherea 
Camenis Attulit in nostrum deposuitque 
sinum’ intended the words to be construed 
‘attulit, deposuitque in nostrum sinum’ 
with a Tibulline postponement of que ; but 
it is also perfectly possible that he or she 
intended ‘in nostrum sinum attulit, de- 
posuitque’ or again ‘in nostrum sinum 
attulit-disposuitque.’ Most probably he or 
she did not think of the matter at all; for 
how else could the words be arranged in a 
pentameter? But suppose Belling and 
Ehwald are right, what then? Have we 
enough of Sulpicia’s verse to be sure she 
was not influenced by a member of her circle 
whom Quintilian places at the head of the 
elegiac poets of Rome? Again I cannot 
believe that iv. 14 is genuine. Fifteen 
years ago I thought I had shown (Journal of 
Philology ix. pp. 280—286) by an elaborate 
argument, condemned by B. apparently at 
second hand, that Tibullus could not have 
been its author; and coming again to it 
after a long lapse of time I can form no other 
conclusion. It is full of coincidences with 
Tibullus ; and yet there is nothing distine- 
tively Tibulline about it. The same 
thoughts are handled and the same language 
used, but in the manner of an inferior artist. 
Thus in 13, 14 alone the composer drew 
from i. 3, 89 sq. and 5, 39 sg.; in 24 fromi. 
5, 58 and 4,71 sg. Besides this the piece 
contains a quantity of verHal and other 
agreements with Propertius, such as is not 
found in the genuine work of Tibullus. 
Repetitions and reminiscences of themselves 
and others occur of course in the work of 
original poets, but not thus congested and 


without the transmuting touch. He who 
borrowed thus wholesale from Tibullus 


and Propertius was not a Tibullus any more 
than he was a Propertius, Set this compo- 
sition aside, and there is not the slightest 
evidence for assigning to him any part of the 
omnium gatherum of the last two books. 
If, as appears most probable, it was 
compiled after the death of Messalla in A.». 
3—by the way Belling’s attempts (pp. 64 
sqq.) to show that in iii. 5, 17-20 Lygdamus 
did not imitate Ovid are most unfortunate 
—twenty years or more had then elapsed 
since the poet’s death and a piece which 
bore the poet’s name might easily have 
found its way into a collection, the last piece 
of which was Domitius Marsus’ epigram on 
his death. Critics would have condemned it 
long ago but for the ascription inv. 13, 


78 THE CLASSICAL REVIEW. 


The device however was not difficult to hit 
upon. Its inventor would have shown more 
penetration had he seen that it is not like 
Tibullus, though it is like Propertius, to 
refer to himself by name, except where, 
as in epitaph (i. 3, 55) or inscription, the 
name is indispensable, and that Tibullus 
has only once used Ahe rhythm of the verse 
which ascribes it to him, in the very place 
(i. 9, 83) from which, as I believe, the 
idea of this ascription was derived. 

Of the guaestiones Tibullianae pp. 3—18 
consists of answers to the objections of 
reviewers (Cr., Magnus, Rothstein), of 
which I have taken account in the foregoing. 
Next follow some corrected statements of 
the reading of A. I may quote i. 3 title 


egrotet (not egi utet) ; 12 omina A ex. corr.; 9, 
19 O wictis; iv. 1, 82 artos; 165 rigentem 
(m sec. ma. in ras, potest fuisse s); iv. 12, 2 
‘wideas’ agreeing with V. The last four 
pages are filled with a transcript of Hein- 
sius’ excerpts of certain fifteenth century 
codices, preserved at Berlin (MS. Diez. B. 
Sant. 55, d.) ‘ut aliquando quid e codice y 
traditum sit, comperiatur.? From the 
transcript it may now be seen how justly 
they have been neglected. In conclusion 
we may express the hope that the author of 
these two pamphlets will not abandon the 
critical investigations for which in many 
respects he appears eminently qualified. 


J. P. PostGate. 





HARTLAND ON THE 


The Legend of Perseus, by E. 8S. Harrianp, 
FS.A., Vol. I. Nutt: London, 1894. 
7s. 6d. 


To review a work of which the first volume 
alone has appeared is a matter of difficulty 
and is not altogether fair to tk» author, 
when, as in this case, the volume contains 
merely the author’s facts and not the con- 
clusions which he thinks they point to. It 
must sufficetherefore now briefly to point 
out the importance and interest of this book. 
As for its importance, enough of Mr. Hart- 
land’s work has even now appeared to show 
that it is one of the most learned contribu- 
tions to folk-lore which has ever been 
produced by an English scholar. The 
volume contains only facts, but facts in 
such abundance that they are a monument 
of industry and learning. The interest of 
the book is threefold. First, it appeals to 
the student of folk-lore: an incident (the 
supernatural birth of the hero) in a world- 
wide story is shown to have its parallel in a 
world-wide custom (that of endeavouring to 
cure sterility by means of drugs, charms, 
etc.). What inference Mr. Hartland will 
draw from this remains to be seen: the 
facts seem enough to show both that the 
incident might have been suggested to any 
number of story-tellers independently, and 
that a tale containing the incident would 


LEGEND OF PERSEUS. 


find everyrhere suitable soil in which to 
take root. Next, the volume appeals to the 
student of apologetics: he will find Mr. 
Hartland quoting from the legends of all coun- 
tries a number of instances of supernatural 
and miraculous conception. The origin of the 
belief enshrined in these legends is apparent 
from Mr. Hartland’s account : the belief in 
the power of curing sterility is world-wide ; 
springing from it is the belief in partheno- 
genesis produced medicinally or miracu- 
lously—this belief is apparently less widely 
spread ; and finally there is the belief that 
heroes, founders of dynasties, and men-gods 
must have differed in their birth, as in their 
life and death, from other men, and there- 
fore must have been miraculously born of a 
virgin—this last belief is apparently the 
least widely spread of the three. What is 
to be inferred from these facts, Mr. Hart- 
land again does not explicitly state. Finally, 
the classical student will be interested in 
Mr. Hartland’s book: the legend of Perseus 
takes him under Mr. Hartland’s guidance 
back to primitive savagery and forward 
through folk-lore to modern times. The 
book, which forms the second volume in a 
new series, entitled the Grimm library, is, 
like all books issued by Mr. Nutt, beauti- 
fully got up. 
ait F, B. JEvons. 








-TOVOS. 


THE CLASSICAL REVIEW. 79 


THE MODES OF 


Tae Editor of this periodical having 
given a courteous welcome to the suggestion 
that I should write something by way of 
reply to the review of my book which 
appeared in fhe December number, I must 
begin by expressing the pleasure with which 
I found myself in the hands of so competent 
acritic. It is not much to say that Mr. 
Stuart Jones has an exceptional mastery of 
the ancient authorities, as well as the 
modern literature, on the subject of Greek 
music: for his competitors are few. But 
however numerous they might be, I venture 
to think that his scholarly pre-eminence 
would be no less marked—éerel ot tu pepty- 
peévos Eotat dpirw. 

At the outset of his article Mr. Stuart 
Jones states the question at issue. My 
object, he says, is to ‘ prove that the modes 
were nothing more than keys differing only 
in pitch, and that of modes in the modern 
sense one only was known to Classical 
Greek music, viz. a scale corresponding to 
the modern minor as played in descending 
from A to A, and represented by the 
standard “ Dorian” octave, extending from 
E to E, but having for its tonic the “ péon”’ 
(A). If the meaning is that I regard this 
seale as the only one recognized and named 
in the earlier Greek music, I should accept 
the statement. I do not however exclude 
the possibility of other scales having been 
in actual use. What I deny is that variety 
of scale or ‘mode’ was the basis of the 
distinction between Dorian, Phrygian and 
the rest. These names were certainly used 
by Aristoxenus to denote keys (révor), and 
by Plato and Aristotle to denote what they 
termed dppovia. In my view dppovia in 
this connexion meant the same thing as 
The prevailing ‘modes’—in the 
modern sense of that term—lI believe to 
have been the Dorian, E to E, and the 
Hypo-dorian A to’ A. Bnt, as I have 
explained on p. 101 of my book, I think it 
possible that other ‘modes’ were known, 
and even that, in the case of the Mixo- 
lydian, the difference of mode was associated 
with a particular key. 

Coming to the passages on which the 
discussion necessarily hinges, Mr. Stuart 
Jones makes a concession which I must 
regard as one of great importance—one 


which separates him from Westphal, and I 


think also from Gevaert, if it does not 
bring him over altogether to my side. He 


GREEK MUSIC. 


admits that ‘ different dppoviat were “@s550- 
ciated with different degrees of pitch’ 

‘that the dpyovia, as usually executed, 
differed in pitch, and fell roughly into three 


groups, viz. the ‘loosely-strung’ or low- 
pitched modes (yaAapai, dveimevat), the 


highly-pitched modes (ovvrovor), and the 
intermediate modes (Dorian and Phrygian) : 
and that the difference in pitch, according 
to Aristotle and Plato, was connected with 
the difference of ethos which characterized 
the several dppoviat. 

With regard to this last point I must 
observe that Mr. Stuart Jones has somewhat 
misconceived the place which it holds in my 
argument. He advances some ingenious 
reasons for thinking that Aristotle and 
Plato were mistaken in connecting differ- 
ence of ethos with difference of pitch,— 
that they were led away by the verbal 
association of the word ovvrovos with the 
notion of ‘highly strung,’ ‘ organized,’ and 
of padakds or dvemévos with ‘relaxed,’ 
‘effeminate,’ and the like. He points out 
also that the Phrygian and Hypo-phrygian 
modes had an orgiastic character which 
their pitch did not account for. In all this 
he may be right. The power of words over 
the scientific imagination is such that even 
Aristotle may have been led astray. But 
the question whether Aristotle and Plato 
were right in their theory of musical ethos 
does not in any way affect my argument. 

What Iam concerned to prove is, not that 
difference of pitch was the source of ethos, 
but that difference of pitch was the chief or 
sole ground of distinction between the ancient 
dppovia. When I dwell on the importance 
which the philosophers attached to pitch, it 
is the fact from which they argue, not the 
correctness of their conclusion, that I seek 
to establish. And in this view the difficulty 
about the Phrygian dppovia is valuable as 
showing that Aristotle, at least, was not 
led away by a general theory to the extent 
of ignoring the plain facts of the case, I 
do not admit, then, that the arguments used 
by Mr. Stuart Jones on the connexion of 
pitch with ethos show that my proof is 
inconclusive. On the other hand I am glad 
to have his support in rejecting all attempts 
to explain ovvrovos and yaAapds or dveipevos, 
in their application to scales, from the order 
of the intervals, or the note on which the 
melody closed (Westphal), or the general 
character of the music. The fact which 


80 THE CLASSICAL REVIEW. 


these terms connote—and in _ ancient 
opinion the all-important fact—is the 
relative pitch of the scales. 

But if the ancient dpyoviat were dis- 
tinguished by pitch, how did they differ 
from the tovo. of Aristoxenus? The rovor 
which that writer specifies as generally 
known in his time bear the same names as 
the dppoviac which we find in Aristotle and 
Plato. And the order, in respect of, pitch, 
is evidently the same. How then can the 
two things, dppovia and rovos, be different ? 
If dppovia implies a certain pitch, that is 
the same thing as saying that it implies a 
particular tovos. 

The argument is illustrated and confirmed 
by the difficulty of explaining why the 
names of the keys and the same names as 
applied to the species of the octave follow 
each other in reversed order. Mr. Stuart 
Jones accepts the reason given by Gevaert, 
viz. that the keys were used in order to 
reduce melodies composed in the various 
modes to a common standard of pitch, e.g. 
the Phrygian species being from D to D, 
and the Dorian from E to E, the Phrygian 
octave must be raised a tone in order to 
agree in pitch with the Dorian. This is in 
fact the basis of Ptolemy’s scheme of modes. 
But when and with what intention was this 
reduction to a common standard of pitch 
brought about? Not, surely, when the 
modes differed in pitch, and were thought 
to derive their character and value from 
that. difference. Not therefore in early 
classical times, but in the time of Ptolemy, 
who deliberately excludes the pitch of 
scales as an element of his scheme. I may 
add that this explanation does not evade the 
force of the argument founded on the un- 
settled condition of the keys in the time of 
Aristoxenus. If the keys were devised in 
order to reduce the modes to a common 
system of any kind, they must have been 
from the first as regular and complete as 
they are in the scheme of Ptolemy. And 
why should Aristoxenus have increased the 
number beyond seven? All this surely 
points to the conclusion that it was not the 
keys but the species of the octave which 
were devised in order to produce a series of 
scales agreeing in pitch and exhibiting all 
the possible successions of intervals. 

Mr. Stuart Jones is surprised (see the 
footnote on p. 452) at my saying that the 
several modes, as represented by the nota- 
tion, could not have followed each other in 
an order the reverse of the order of the 
keys. The actual scales of the modes, as 
he admits, followed the same order as the 


keys. He says that the scales in question 
were not necessarily the actual scales 
employed. I cannot understand how the 
scales of the modes as given by the notation 
could have been other than those actually 
employed. 

Regarding the evidence of Plutarch, 
which Mr. Stuart Jones ranks as the second 
of my chief arguments, not much need be 
said. In the dialogue on Music, as I 
showed, he uses rovos and dppovia promis- 
cuously: in a passage of another work, 
quoted by Mr. Sandys, he expressly says 
that the words mean the same thing. Mr. 
Stuart Jones finds the statements of Plutarch 
‘somewhat confused.’ If he is wrong in 
identifying tdvos and dppovia, his statements 
are more than confused. If he is right, he 
can only be accused of using two words 
when one would have sufficed. It is true 
that Plutarch is not an accurate writer, and 
is much too late to carry decisive weight. 
On the other hand he shows great know- 
ledge of earlier writers, especially of 
Aristoxenus, from whom much of the 
dialogue on Music appears to be borrowed. 

I turn now to the ‘ positive indications’ 
which Mr. Stuart Jones puts forward in 
support of his view. The first of these he 
finds in the Philebus (p. 17), in a passage 
which perhaps ought to have been discussed 
in company with the other Platonic texts. 
In this passage Plato speaks of the intervals 
(S:acrHpara) as the elements which by com- 
bination produce scales (cvarjpata), known 
to musical tradition as dppovia. From this 
Mr. Stuart Jones draws the acute inference 
that the word which replaced dpyovia in the 
sense of ‘mode’ was not rovos but cvornpa. 
Thus the dppoviac of Plato (and Aristotle) 
are the seven octachord ovorjpata which 
Aristoxenus speaks of as having been called 
dppoviat by the ancients. 

Plausible as this is, I am unable to 
reconcile it with the doctrine of the ovor7- 
pata as laid down by Aristoxenus and his 
followers, or with what we know of the 
scales actually employed. If the old 
éppoviat (Dorian, Phrygian, &c.) are not the 
rovo. but the ovorjpara of Aristoxenus, 
why do we never hear of a Awpiov cvornpa, 
a @pvywov ovoryua, and so forth? On the 
contrary, the only distinctions in the ovarn- 
para are based either upon their compass, 
or upon the difference between conjunct and 
disjunct tetrachords. I need not add that 
the notion of a particular pitch, which was 
at least practically involved in the dppovia, 
is always absent from the description of a 
ovoTna. Again, how comes it that the 


















THE CLASSICAL REVIEW. 


ovoTna or succession of intervals given 
by the notes of the octave from 
to vit is always the same in 
genus? Why is it always an _ octave 
of a particular species? Why, in short, is 
such a thing as a standard or ‘perfect’ 
system recognized in Greek music? No 
one mode had any such vogue or superiority 
to the others. I conclude therefore that 
the ovorynpara in the Philebus are not 
‘modes,’ but either simply the keys (as a 
friend has suggested to me), or the varieties 
of scale given by the different genera and 
‘colours,’ as well as by the option between 
disjunct and conjunct tetrachords. 

The next piece of evidence is a passage 
of Plutarch in which he says that it was 
discovered by a certain Lamprocles that in 
the Mixolydian mode the rovos d:alevnrixds 
was the highest interval, so that the suc- 
cession of intervals was as from B to B 
(iratn tratav to wapapéeocn). This state- 
ment I endeavoured to ‘minimize,’ 7.e. to 
ascertain exactly how much it proves. I 
will not now insist on the circumstance 
that it comes to us from an author who 
expressly tells us that révos and dppovia 
mean the same thing: for Plutarch is 
admittedly inaccurate. The important 


brary 
each 


point seems to me to be that this M 
lydian octave is obtained by a change of 
system such as the ancient authorities 
recognized, viz. from disjunct to conjunct. 
In the ‘Dorian’ octave E—E substitute 
for the notes A BC D the conjunct tetra- 
chord A BoC b, and the octave becomes 
*‘Mixolydian.’” Thus the Mixolydian is a 
scale which is provided for in the tetra 
chords of the Perfect System: whereas the 
Phrygian and Lydian octaves, taken on 
that System, are not bounded by ‘ standing’ 
notes. This is my reason for doubting 
whether we can argue from the Mixolydian 
to the other modes. : 

With regard to the passage of Aristides 
Quintilianus on the dppovia of the Republic 
I have only to say that I do not trace it to 
the same source as the scale xara ddoas of 
the same author. My argument is merely 
that if the latter, which claims high 
antiquity (zapa rots dpyxatos), is found to be 
a forgery, some degree of suspicion is cast 
on the former also. I am glad that Mr. 
Stuart Jones agrees with me as to the 
character of the scale xara does. Previous 
commentators have treated it as a genuine 
document. 

D. B. Morro. 


STATIUS, SIZV. I. vt. 44. 


Una uescimur omnis ordo mensa, 
Parui femina, plebs eques senatus. 


Here parui pl. adjective beside /femina 
sing. substantive is strange ; and the anti- 
thesis required to woman is rather man 
than children. Baehrens reads mas et femina. 
A much simpler alteration would be 


por uir femina, plebs eques senatus. 


Cp. Ov. 7rist. I. iii. 23 femina uirque meo, 
pueri quoque funere maerent. 
J. S. PHcuimmore. 


[Though this correction appears to me 
certain, parui might derive some support 
from Lucan ii. 108 crimine quo parut caedem 
potuere mereri! 

8S. G. Owen, | 


SIR C. T. NEWTON. 


[An Address delivered at the General Meeting of the Hellenic Society on Jan, 23, 1895, by 
Prof. R. C. Jebb, M.P., President of the Society. | 


Av the first General Meeting of this 


Society which has been held since the death 


of Sir Charles Newton, it is fitting that some 
tribute should be rendered to the memory of 


one whom the Hellenic Society may justly 


regard as chief among its founders ; whose 
_ NO. LXXy. VOL. IX. 


presence and influence did more than any- 

thing else to carry it successfully through 

the earliest days of its existence ; and who, 

to the end of his life, took the keenest in- 

terest inits growing prosperity. It is fitting 

also that we should recall to-day, at least in 
G 


82 THE CLASSICAL REVIEW 


outline, the salient characteristics of the 
distinguished career to which our Society 
owes so large a debt. 

Newton’s life divides itself into three well- 
marked chapters. The first contains the 
thirty-six years from his birth in 1816 to 
1852 ; it is the period of preparative studies. 
The second begins in 1852 with his consul- 
ship at Mitylene, and closes in 1861 with his 
return to London as the head of his 
Department at the British Museum ; it com- 
prises the period of travel and discovery 10 
the Levant. In the third chapter, from 1861 
onwards, he is the organizer and admin- 
istrator ; the recognized head of classical 
archaeology in this country ; the active sup- 
porter of all enterprises, whether originating 
at home or abroad, which could extend the 
knowledge of antiquity, or which promised 
to advance an object always so near to his 
heart, the addition of new treasures to our 
great national collection. 

From Shrewsbury School, then ruled by 
that brilliant scholar, Samuel Butler, 
Newton ‘went in 1833 to Christ Church, 
Oxford, where he attracted the favourable no- 
tice, and strongly felt the influence, of Dean 
Gaisford. He was also for a time the pupil 
of his lifelong friend, Dean Liddell. Mr. 
Ruskin, who was an undergraduate member 
of the House at the same time, has recorded 
in Praeterita the particular trait which most 
impressed him in Newton ; it is one which 
can be easily recognized by those who knew 
him in later years—‘ his intense and curious 
way of looking at things.’ 

In May, 1840, Newton became Junior 
Assistant in the Department of Antiquities 
at the British Museum. That Department, 
founded in 1807, was not then constituted 
as it isnow. In 1861 it was subdivided into 
three provinces ; Greek and Roman Antiqui- 
ties ; Coins ; and a third, in which Oriental 
Antiquities were associated with British and 
Mediaeval; the two latter, with Ethno- 
graphical Antiquities, were detached from 
the Oriental in 1866. But, in 1840, the 
opportunities which Newton found at the 
Museum, if less adapted to the training of a 
specialist, were well suited to encourage a 
comprehensive view of antiquity. At the 
head of the Department was Edward 
Hawkins, a man of varied attainments, but 
especially a numismatist ; and Newton's 
early studies in that direction left on his 
mind the conviction that numismatics, be- 
sides their special interest, have the highest 
value as a general introduction to classical 
archaeology. 

Among his earliest publications, there is 


one which has a peculiar interest. In 1847 
he wrote a paper on some sculptures from 
Halicarnassus—they were, in fact, parts of 
the frieze of the Mausoleum—which had 
lately been secured for the British Museum 
by Sir Stratford Canning. In this memoir, 
Newton conjecturally placed the Mausoleum 
in the centre of the town of Budrum, from 
the fortress of which the above-mentioned 
sculptures had come. A description of the 
site by the architect Donaldson—confirming 
the account by Vitruvius—pointed to this 
conclusion. Ten years later he was to prove 
its truth. Such competent explorers as 
Spratt and Ross, misled by the appearance 
of the ground, had looked elsewhere. 

In 1852 Newton, whose qualities were 
becoming well known, was appointed Vice- 
Consul at Mitylene. It was in reality, 


though not in form, an archaeological mis- 


sion. Lord Granville, then Foreign 
Secretary, was doubtless well acquainted 
with the new Vice-Consul’s gifts. New- 
ton had able assistance in the routine 
duties of the post. From April, 1853, 
to January, 1854, he was at Rhodes, 
and thus within easy reach of the region 
in which his chief work was to be done. 
The six years which followed were rich in 
results. He explored the island of Calymna, 
off the Carian coast, and obtained some 
remains of early Greek art which are now 
in the room of Archaic Sculpture at the 
Museum. At Cnidus, in a sanctuary of 
Chthonian deities, he found the beautiful 
seated statue of Demeter, in which Brunn 
recognized the perfect ideal of the goddess. 
Among other monuments discovered at 
Cnidus is the lion, supposed to commemorate 
Conon’s victory in 394 B.c. From Bran- 
chidae, near Miletus, Newton brought away, 
besides a lion and a sphinx, ten archaic 
statues of seated figures which had stood by 
the Sacred Way leading from the temple of 
Apollo to the harbour, It was under a 
firman which he procured that the bronze 
serpent at Constantinople, inscribed with 
the names of the Greek cities allied against 
Xerxes, was first disengaged from the 
soil; though the task of deciphering the 
inscription was reserved for Frick and 
Dethier. 

But his most signal achievement was in 
connexion with the Mausoleum at Halicar- 
nassus. It was in 1855 that he first saw 
the castle of Budrum, and found fragments 
of sculpture embedded in its walls. Lord 
Stratford de Redcliffe, then British Ambas- 
sador at Constantinople, who had constantly 
supported Newton in all his work, promised 





























THE CLASSICAL REVIEW. 8. 


to obtain the necessary firmans. In the 
autumn of 1856 Newton visited London, 
and, aided by Sir Anthony Panizzi, Principal 
Librarian of the Museum, secured the 
assistance of Lord Clarendon, who was then 
Foreign Secretary. A ship of war was 
placed at his disposal, with a party of men 
of the Royal Engineers, under the command 
of the officer who is now General Sir R, 
Murdoch Smith. On Jan. 1, 1857, Newton 
broke ground at Budrum. The sculptures 
with which that enterprise enriched the 
Museum are, for the fourth century B.c., 
almost what the Elgin marbles are for the 
fifth; as the latter illustrate the art of 
Pheidias and his school, the remains of the 
Mausoleum throw a comparable light on the 
art of Scopas. Indeed, it was Newton who, 
both by his discoveries and by his penetrating 
analysis, opened a new era in the modern 
knowledge of that sculptor. 

In May, 1860, Newton was appointed 
Consul at Rome. But he stayed there only 
about a year. The reconstitution of the 
Antiquarian Department at the British 
Museum was a measure to which the wealth 
of Newton’s acquisitions bad mainly con- 
tributed ; and nothing could be more appro- 
priate than that, when a separate Depart- 
ment of Greek and Roman Antiquities was 
created in 1861, he should be invited to 
preside over it. 

The earliest years of his new office were 
marked by the publication of those two 
books which record his work in the Levant. 
In 1862 appeared his History of Discoveries 
at Halicarnassus, Cnidus, and Branchidae ; 
it is essentially a scientific work, addressed 
to experts. Three years later came the 
Travels and Discoveries in the Levant ; a book 
profoundly interesting to all students, but 
also with a popular side ; it has been well 
described as ‘a charming Odyssey,’ enlivened 
with pictures of Greek and Turkish man- 
ners,—lit up, indeed, with all the colours 
and humours of Anatolia, such as it was 
half-a-century ago. This work, admirably 
illustrated, owed not a little of its charm 
to the pencil of the accomplished lady who, 
a few years before, had become the author’s 
wife; a daughter of the Joseph Severn 
whose grave at Rome is beside that of his 
friend John Keats. One year later, in 1866, 
the crushing sorrow of her death befell 
Newton ; and the shadow of that loss never 
passed away. 

Newton held his post at the Museum for 
twenty-four years—till 1885. His activity 
during that period has two principal aspects ; 
one, directly relative to his oflice itself ; the 


other, relative to the influence and position 
which that office conferred. 

As keeper of the Greek and Roman 
Antiquities, he rightly felt that, next to the 
duty of organizing and conserving those 
treasures, his first duty was to augment 
them. Here his social and diplomatic 
wbility, joined to the prestige of his dis 
coveries, gave him unique advantages. In 
the first three years of his tenure, the 
annual grant from the Treasury for pur 
chases in his Department rose from £785 in 
1861 to £1400 in 1864. In the twenty 
years from 1864 to 1883, a series of Special 
Parliamentary Grants, amounting in the 
aggregate to about £100,000, enabled him 
to secure for the Museum objects of first- 
rate importance in every branch of archaeo 
logy, including the choicest things of all 
sorts in four inestimable collections,—the 
Farnese, the Pourtalés, the Blacas, and the 
Castellani. 

This was one side of his energy,—that 
immediately connected with his function at 
the Museum. But, in virtue of his position 
and influence, he was also enabled to stimu- 
late and assist research in every quarter of 
the classical lands. It was thus that he 
furthered the work of Messrs. Smith and 
Porcher at Cyrene; of Mr. Wood at 
Ephesus ; of Mr. Pullan at Priene ; of Mr. 
Dennis at Benghazi in Tripoli: and of 
Messrs. Salzmann and Biliotti in supplemen- 
tary researches on the ground which he had 
made his own, at Budrum. 

When the inaugural meeting of this 
Society was held, in June, 1879, it was to 
him that the supporters of the project 
primarily turned for countenance and 
counsel. During the first six years of the 
Society’s life, he was constantly in the chair 
at our meetings ; nor is it too much to say 
that his guidance and his name must be 
reckoned among the chief causes, not only 
of the early and rapid success which 
attended the Hellenic Society, but also of 
the position in which it is now established. 
In 1883 his aid and counsel were also 
valuable in helping to institute that British 
School at Athens which, in the face of 
difficulties not experienced by the similar 
schools of other nations, has done so much 
to uphold the reputation of our country in 
the tield of archaeological research. — 

This is merely a bare outline of Newton's 
life-work ; but even so slight a sketch must 
not close without some attempt to indicate 
the leading characteristics of the man’s 
mind and nature. First, as to his attitude 
towards his chosen studies. It has lately 


G2 


84 THE GLASSICAL REVIEW. 


been said, by one well fitted to judge, that 
the ancient monuments interested Newton 
rather on the historical side than on the 
mythological or the artistic. Indeed, his own 
words can be quoted ; ‘IT am first a historian, 
and secondly an archaeologist.’ This may 
seem a hard saying; but I believe that it is 
true, though it perhaps needs some elucida- 
tion. It means that Newton was never a 
specialist in the limited modern sense ; it 
was classical antiquity as a whole that had 
a spell for him ; it was in the intense desire 
to reconstruct and revivify this antiquity 
that he so closely and indefatigably scanned 
every monument of any kind that could tell 
him anything about it. His strongest feeling 
in early manhood was that ancient literature, 
in which he was well versed, told only part 
of the story. His address at Oxford in 
1850—which now stands first in his volume 
of Essays—begins with words which strike 
the key-note of his work: ‘The record of 
the human past is not all contained in 
printed books.’ Hence the peculiar interest 
which he always took in epigraphy ; here 
he felt that he came closest to ancient lives 
and minds: his two essays on Greek inscrip- 
tions, published in 1876 and 1878 (the 
fourth and fifth of the collected Essays), 
illustrate this in full; few productions of 
his pen are more striking. 

Now, this desire to apprehend the life of 
antiquity is often associated with the kind 
of imagination which seeks vivid or rhe- 
torical utterance in language; it was 
distinctive of Newton that, in his case, there 
was absolutely no such tendency ; on the 
contrary he recoiled from it. The life of his 
imagination was an inward life,—so inward, 
that he might often seem unimaginative ; 
a life which he wished to share only 
with the careful, laborious, exact student, 
but did not choose to share with the outward 
world. Witness the guides which he prepared 
to his galleries at the British Museum— 
exemplifying his conception of a_ scientific 
catalogue as the outcome of a life devoted 
to a single study—but making no concession 
to a popular desire for more elementary 
knowledge. When, in 1880, he became the 
first Professor of Archaeology at University 
College, London, the stamp of his teaching 
showed the same bent. 

His sustained, though undemonstrative, 
ardour was singularly allied with caution. 
Without being cynical, he was wary in a 
degree which sometimes approached to 
cynicism ; in discriminating between what 
was merely probable, and what might be 
accounted certain, he leaned to the sceptical 


side ; and he was imbued with the sentiment 
which Aristotle attributes to old age, that 
‘most things are unsatisfactory.’ No man 
was less sanguine, or quicker to foresee the 
difficulties of a project ; but, once engaged 
in it, he was tenacious and intrepid. His 
self-contained manner was due in part to 
the natural fastidiousness .of his taste; it 
was only when he felt secure, for the time, 
against jarring incidents,—which, even 
when slight, affected him like physical pain, 
—that he completely unbent, and showed 
the most genial side of his nature. In 
colleagues he looked forthe highest standard 
of work; his demands on subordinates were 
strict: he was an exacting, but also a 
stimulating ruler. 

If the essence of his character could be 
contained in a phrase, it might perhaps be 
described as severe enthusiasm. To those 
who knew him but slightly, the severity— 
not harshness, but the severity as of good 
Greek sculpture—might be more evident 
than the enthusiasm: but a nearer know- 
ledge revealed the man in whom an inward 
fire had burned steadily from youth upwards ; 
a sacred fire, little seen, but not to be extin- 
tinguished, and shaken neither by any 
wavering of purpose, nor by the breath of 
any vulgar ambition. His many honours, 
academic or public, were prized by him in 
proportion as he took them to be recogni- 
tions, not merely of eminence generally, but 
of success in the precise aims which he had 
set before himself. 

The chief source of satisfaction to him, 
in his later years, was to think that classical 
archaeology had gained so much ground in 
England, and that he had helped it forward ; 
but this feeling was deeply tinged with 
melancholy ; he thought of himself as the 
leader through the wilderness, who was not 
to enter the promised land. There are 
minds, perhaps, in which life-long conversa- 
tion with the past so confirms the habit of 
retrospect that the difficulties of earlier 
years always loom large, even after subse- 
quent successes ; So, at least, it seemed to 
be with him. But to others it will appear 
that, however distant the point gained in 
his lifetime may have been from his ideal, 
still the cause to which he rendered such 
abundant service was already gained before 
he died. In the future of classical studies, 
so long as they may exist in this country, 
the place of archaeology, not as an accident 
but as an essential, is assured beyond the 
danger of overthrow. 

Newton has been recently compared, and 
not unjustly, with Winckelmann. The 





THE CLASSICAL REVIEW. RE 


German worked in the dawn, the English- 
man, though still in the morning hours, yet 
in a far clearer light: between them, 
however, there is this intrinsic resemblance 
that in both the mainspring of a devotion 
which ended only with life was a native 
instinct, intensely strong and lucid, for the 
spirit and the charm of classical antiquity. 
There are those in this room to whom the 
impressive personality of the master whom 
we commemorate will be a lasting recollec- 
tion,—that singularly fine head and_ pose, 
which themselves seemed to announce some 


kinship with ancient Hellas.—that 
which so often within these walls expressed 
the knowledge thrice-refined by ripe study 
and experience ; a few years more, and these 
will be only traditions: but to our sue 
cessors, the members of this Society in days 
to come, the history of learning in Europe 
will bear witness that no body formed for 
the promotion of Hellenic studies could 
have entered upon existence with a worthier 
sanction, or could desire better auspices for 
its future, than those which are afforded by 
the name of Charles Newton. j 


Voice 


IN MEMORIAM : CHARLES THOMAS NEWTON, K.C.B. 


DrEcEMBER 47H, 1894. 


ov Kw podoov eros: ypvoavyéos avOean roins 

Tata zpodpacca deper xepepiv’ oiyonevw. 

[ ov yap azdvOpwros, daxpvwv aéxnte wobewar, 
és paxapwov Néecxnv oidGev olos are. 


















SOME POINTS IN DR. FURT- 
WAENGLER’S THEORIES ON THE 
PARTHENON AND ITS MARBLES.! 


Dr. Furrwaneer’s Meisterwerke der 
griechischen Plastik was very fully reviewed, 
and indeed its contents were summarized, in 
_ the Classical Review for April and May of the 

past year. The object of the present article 
is to call attention to the appearance of an 
English translation of the work, and to use 
this occasion for the criticism of certain 
points in Dr. Furtwiingler’s theories on the 
Parthenon marbles. I may say at once 
that the work of editing the English version 
1 Masterpieces of Greck Sculpture, a series of Essays 


on the History of Art. By Adolf Furtwiingler. 
Authorized Translation. Edited by Eugénie Sellers. 


- 
+ 


ovK, GAN’ ovAos e€pws Tains o wdwoe tporoprod, 
Kal puopos Kpadinv eixe AGov Lapins: 

dotis éverkAnkns ipov te kal’ “EXAddos ofdas 
kat ottBéwv ’Acins wy’y.ov TEnEVoS, 

évOev 6 Mavowdos rpocpednocaca te Anu 
got KX€os abavatov Tice Kopiotpa Ged. 

ei 6€ ceoiyntar PoiBos dadvar te papavber, 
Tappeve pavtevowy THE piAovoe diAos. 


Greorce C. W. Warr. 


ARCHAEOLOGY. 


has been admirably done by Miss Sellers- 
She has undertaken it as a confessed en 
thusiast, accepting her author's views em 
bloc ; the translation is all aglow with eager 
championship, and indeed only the devotion 
of an ardent disciple could have carried her 
with such brilliant success through a task 
veritably Herculean, The English transla- 
tion will, I expect, largely supersede the 
German version, even in Germany, It is in 
almost every respect a gain. The changes 
made are noticed by Miss Sellers in her 
preface, and have all been authorized by 
Dr. Furtwiingler. The plates are now in- 
corporated with the text, and there are no 
less than thirty-five new illustrations; yet 
the bulk of the book is not seriously in- 
creased. The necessary space has been 


86 THE CLASSICAL REVIEW. 


gained by condensations freely and wisely 
carried out. Current controversies of minor 
importance have been relegated to notes, 
The book has gained in unity by the omis- 
sion of chapters that dealt with pre-Pheidian 
art; in one instance, 7.e. the chapter on the 
throne of Apollo at Amyclae, we regret the 
omission, though for consistency’s sake it 
was justifiable. 

The translation is for the most part clear 
and accurate ; but Germanisms abound and 
sometimes issue in obscurity, e.g. Preface, 
p- Viil.: ‘the increasingly rich discoveries of 
original work on Greek soil have lately 
somewhat thrown into the shade the study 
of the copies for which we are mainly in- 
debted to Italy, not to the advantage of our 
science.’ ‘Hs ist die Auswahl des Besten 
und Beriihmtesten das man im Altertum 
besass’ is German. ‘It is the pick of the 
best and most famous that antiquity pos- 
sesses’ would be written independently by 
no Englishman. ‘The no less magnificent 
bronze head of a boy’ is a severe ‘split 
substantive’ for an English tongue. As 
a nation we are still so far behind in archaeo- 
logy that many of us all but think in 
German on the subject, but a stern set must 
be made against writing Germanized English. 

I turn now to the immediate subject of 
this paper, Dr. Furtwingler’s views on the 
Parthenon; and specially its marbles. With 
more enthusiasm than accuracy Miss Sellers 
says in her preface that Dr. Furtwiingler’s 
book has been received ‘almost with ac- 
clamation by scholars of all schools.’ As 
regards the Parthenon marbles it is surely 
better to own frankly that his theories have 
been met in England not with acclamation 
either way, but with a grave distrust. 
I myself believe and gratefully acknowledge 
that Dr. Furtwiingler has thrown brilliant 
light on the whole question of interpreta- 
tion, but in certain serious matters of 
detail I am at issue with his conclusions, 
and these for convenience sake may be stated 
at once and together. 

(1) In the West pediment the ephebos 
called Erysichthon by Dr. Furtwangler I 
believe to be Erichthonios. 

(2) In the East pediment the so-called 
‘Theseus’ is not Kephalos. 

(3) In the central group of the East 
frieze the scene represented is not the offer- 
ing of the peplos. 

(1) The West pediment.—Dr. Furtwiingler 
rightly sees in the centre group not a strife 
but a rival theophany — scarcely perhaps 
even rival, for by the time of Pheidias 
Athene and Erechtheus were conjointly 


worshipped, not in the Erechtheion but in 
the ‘alte Tempel.’ Of this more anon, One 
mythological comment may be added. As the 
medium of their theophany we see their char- 
acteristic onpeta. Athene shows the olive, 
Poseidon the horse. We go a step further. 
The gods in their theophany show them- 
selves, z.e. Athene is the olive, Poseidon is 
the horse —“Immios; the importance for 
Athene will appear later in connection with 
the Moirae ; for the present it is essential to 
note that the horse to our mind was the 
original onpetov of the god. It does not 
appear in the. pediment—why? Because 
(and this is a point that Dr. Furtwingler 
seems not clearly to have seized) Poseidon 
appears in the pediment, not as Poseidon 
pure and simple, but as Poseidon Atticized, 
i.e. combined with Erechtheus. ‘The old 
primitive worship of Athene knew of certain 
local figures only, the olive-goddess Athene, 
her snake-husband Kekrops, their child Eri- 
chthonios Erechtheus, and the well that 
nourished the olive-tree, called after him 
Erechtheis. Erechtheus is but the snake- 
child Erichthonios conceived as a grown-up 
prince and ancestor. To this primitive triad 
of father, mother, and child, to the old 
snake and the young snake and the olive- 
tree nourished by its sacred well! there 
entered, probably in all but historical times, 
the horse-god Poseidon. He forced his way 
into the cults of the Acropolis, but only by 
suffering assimilation with the local hero, 
the original child-god Erechtheus Erichthon- 
ios ; and more than that, he had to drop his 
horse form (though it lived on in current 
story and in the horse-tribe of Centaurs) and 
take for his symbol the sacred well of Erech- 
theus, originally only important as watering 
Athene’s olive. Given over to him asa sea-god 
in the Olympian circle, it could easily, when 
necessary, be salted for the edification of the 
faithful. Poseidon showing the horse would 
have represented a far less perfect fusion, 
though a poet, like Sophocles in the Colonos 
chorus, may make a splendid conjoint image 
of the white sea-cavalry. 

Similarly the whole right-hand side of the 
pediment is given to Poseidon’s family and 
following, but carefully Lrechtheusized. 
Oreithyia and her Thracian sons are purely 
Poseidonian—Hippian we might say—but 

1 In this matter of the primitive olive, well and 
snake cult, and the consequent close connection of 
Athene and Kekrops—a point I believe to be essential 
—I should like to express my obligations to a paper 
by Mr. A. G. Bather, which, I regret to say, is un- 
published ; also in the matter of the lamp of Kal- 


limachos as representing the hearth of the state ; but 
for the deductions I draw he is in no way responsible, 









































PPS ol 


THE CLASSICAL REVIEW. g 


she, with Ion, Creousa, and the rest, are 
turned into Erechthean princesses; the 
daughters of Erechtheus are nonentities 
invented to acclimatize these foreign women 
—the boldest myth-maker shrank from 
making them Kekropidae. 

Turning to the Kekrops side, from which 
indeed all certainty of interpretation must 
spring, one point seems to escape Dr. Furt- 
wingler, though early commentators (Weber, 
Overbeck, cc.) had noted it, and that is 
that the three women-figures next Kekrops 
are not his three daughters, but his wife and 
two daughters. Dr. Furtwiingler says ‘die 
letzte, wohl die jiingste, kniet neben ihm und 
hilt ihn umschlungen.’ Surely this is to 
mistake the situation. It is Aglauros, wife 
of Kekrops, who clings to him, and the pose 
indicates the relation ; her concern is with 
her husband, the maidens have another 
charge, the boy. Dr. Furtwiingler does not 
seem to have any clear conception of the 
origin of this group of three, who under 
other names appear as Horae, Charites, 
Moirae, and the like. A right understanding 
of these is essential to the interpretation of 
the pediment. MHerse, as I have tried to 
show elsewhere (Hellenic Journal, vol. xii. 
p- 351, ‘The Three Daughters of Kekrops’), 
is merely eponymous of the Hersephoria 
she has no real cult, no shrine, or even pre- 
cinct. Pandrosos and Aglauros are both 
forms—Horae—of the same goddess, who at 
some time or another, perhaps through the 
making of a permanent image, perhaps 





’ through formulation due to the incoming of 


the Poseidon cult, became unified as Athene; 
Athene, in process of time, separated off as 
a distinct superior goddess with her two 
handmaids, Aglauros and Pandrosos. Such 
is the idea in the terracotta cited by Dr. 
Furtwiingler himself, where the seated god- 
dess is attended by two standing Parthenoi 
(Gerhard, Ges. akad. Abh. Taf. 22,1). This 
idea of two and one, not three, lives on even 
in the groups of the Nymphs and Charites, 
where either the middle one faces round or 
sometimes the first of the group is heavily 
draped with a himation for distinction. 
It is most clearly seen in the West pediment, 
where Aglauros the wife, who gave her name 
to the ’AyAavpides, clings to her husband. — 
But I come to what seems to me a crucial 
mistake, Erysichthon. To call the ephebos 
between the two maidens Erysichthon is to 
me to make nonsense of the whole composi- 
tion. If the Delos, Triopas hero was wanted 
anywhere it is on the opposite side. ‘To talk 


of his balancing Ion on the right half of the 


pediment because of his relations to Ionian 


Delos is merely to darken counsel, If he j 
Ionian let him go over and stand near Io: 
Sut the case is much worse than thi Dy 
Zielenski, in a monograph which Dr. Furt 
wiingler does not even cite,’ has, we think, 
demonstrated that the very lix ing Triopi ul 
Erysichthon is the same as the rather 
shadowy Athenian hero, and in Triopas wi 
learn his real nature, 7.e. that he is none other 
than Poseidon Halirrhothios, he who in 
Triopas attacks the sacred tree of Demeter, 
who at Athens tries to cut down the popias, 
the sacred olives of Athene, or, when the 
story is put in human form (nymph and 
tree being, as we shall see, one and the 
same), to violate the daughter of Aglauros, 
Alkippe. It is true that in later times, 
when everything was Atticized, Erysichthon 
was tacked on to the Aglaurides as the 
young brother who died childless. But is a 
figure, in this relation so fatuous, and in his 
real essence so Poseidonian, on the Athene 
side of the pediment to upset that very 
principle of interpretation which Dr. Furt- 
wingler himself so ably maintains? 

I have spoken strongly about Erysi- 
chthon because he has been treated with much 
detrimental vagueness as a ‘shadowy per- 
sonality,’ who can be thrown in anywhere 
when convenient ; and it is time this sort of 
thing stopped, at least among serious mytho- 
logists. 

Before leaving the West pediment a word 
as to Kekrops, rather by way of clearing up 
his mythology than of criticism of the 
pediment. The personality of Kekrops as 
the snake-father god has been much ob- 
scured by the intrusion of the Olympian 
conception of Zeus. As the snake father 
of the child Erichthonios he appears on 
the familiar Berlin terracotta, in his right 
place, balancing the real mother Athene 
Aglauros, all the notions and legends of hes 
being merely foster-mother having arisen 
when the doctrine of Athene’s virginity was 
promulgated at Athens. To explain him as 
an interested spectator only, is to miss the 
point. Even after, as in the Corneto vase, 
the putative father Hephaistos is introduced, 
he takes a back place in the scene, and 
Kekrops holds his own. On the archaic 
poros pediment he holds the eagle in his 
hand—he is Zeus. This notion is confirmed 
by the recent investigations of Dr. Robert ° 
on the worship of the snake child Zeus 
Sosipolis at Crete and Olympia, and, as we 
know from the inscription newly found at 
Magnesia, in that city also. On Magnesian 

1 Philol. N.F. iv. p. 161. 
2 Mittheilungen, 1893, p. 39. 


88 THE CLASSICAL REVIEW. 


coins the snake, the child, and the chest! 
appear in conjunction just as in the Eri- 
chthonios story (Rayet, Golfe Latnuque, p. 
139 ; A. Anzeiger, 1894, p. 81). The analogy, 
indeed, between the cultus conditions of 
Athens and Magnesia is close, as has 
been already indicated by Professor Kekulé 
(op. cit.). We have Zeus Sosipolis = Kekrops 
worshipped with the sacrifice of a chosen 
bull, dpyouévov ordpov, a ceremony that must 
recall the Bouphonia of the primitive Athe- 
nian Zeus with its close Cretan analogies. 
We have a maiden goddess worshipped side 
by side with him for whom an actual Par- 
thenon was prepared. Since we know that 
Themistocles introduced at Magnesia the 
Panathenaia as well as part of the Anthe- 
steria, we may hope from the imported 
ritual to learn some particulars, otherwise 
lost, of the mother-cult. Yet another point 
about Kekrops, and one, to my mind, of 
extreme importance. The snake-god—call 
him Kekrops or Zeus as you will—is the 
father-god, the generating spirit, husband of 
the mother-goddess Athene Agraulos, the 
olive-tree round which the snake is twined. 
He stands on the Athene side of the pedi- 
ment ; his snake, to the end crouched be- 
neath her shield, was tended by her priestess. 
Kekrops, in any distribution of cults and 
division of temple-chambers, must never be 
separated from Athene, must never go over 
to the temple or temple-division of Poseidon ; 
we cannot separate him from the snake, 
and the snake clings always to Athene. 
I heartily agree with Dr. Furtwiingler’s 
dogma that the double back-to-back temple 
denotes the double cult, but it is always 
the cult of affiliated divinities, not of 
those essentially and primarily related— 
husband and wife need no separate shrines 
—-nor yet the snake-child. Athene and 
Poseidon Erechtheus may be and were wor- 
shipped back to back, but Athene and 
Kekrops we may not put asunder. Until 
Poseidon came and made complications these 
back-to-back temples could not and would 
not have existed. To put Kekrops in the 
west hall of the ‘alte Tempel,’ aloof from 
Athene and closely allied with the succes- 
sively immigrant gods Poseidon, Butes and 
Hephaistos, is to introduce an element of 
mythological confusion truly deplorable. 
Dr. Furtwingler’s elaborate distribution 


1 The coin should be studied not in Rayet’s wholly 
inadequate and inaccurate reproduction but in the pho- 
totype plate of Imhoof-Blumer, Griechische Miinzen, 
pl. viii. 33 where the object is manifestly a cista 
with round lid. A similar Magnesian type shows 
the child seated on the cista. 


of cults in the several chambers of the ‘alte 
Tempel’ rests on the analogy he supposes— 
and I think probably with reason — to 
exist between its ground-plan and that of 
the Erechtheion. The account given of the 
Erechtheion by Pausanias can on this sup- 
position be applied to the ‘alte Tempel.’ It 
does not escape Dr. Furtwiingler that Pau- 
sanias makes no mention there of a Kek- 
ropion: ‘ Pausanias erwahnt das Kekropion 
gar nicht, was bei dem Vielen das er tibergeht 
uns nicht wundern darf.’ Pausanias had a 
much more substantial reason for his omis- 
sion. He describes no Kekropion in the 
Erechtheion because there was none to de- 
scribe. But what of the inscription? what 
of the éri ri mpoorace the mpos TO[L] Kex- 
powior? Dr. Furtwingler quite rightly 
observes that in this inscription zpos with 
the genitive indicates direction, ‘nach’ ; 
that is true enough and every one knows it, 
but he goes on roundly to assert that pds 
with the dative must be taken to mean 
contact, mpocracis mpos TO Kexpo7iw, ‘die an 
das Kekropion gefiigte Halle’; and this for 
no better reason than that once in the same 
inscription zpds with the dative does imply 
contact. Now surely this is a wrong and 
most misleading inference. The real dis- 
tinction in the use of the preposition with 
genitive and dative is that the genitive 
denotes direction—the point of the compass 
(of course without motion towards) ; pds 
with the dative denotes proximity with no 
indication of direction, but proximity of 
every varying degree from mere nearness to 
absolute contact. Because contact is ex- 
pressed once in the inscription it does not 
follow that contact is always intended. The 
real grammatical distinction lies, I repeat, 
between direction without implied proximity 
and proximity without implied direction, and 
Dr. Furtwiingler’s argument is based on a 
confusion of grammatical thought. What 
and where, then, is the Kekropion? A 
building near, uncomfortably close up to the 
portico—incompatibly close up, some say. 
The Kekropion is none other to my mind 
than the East cella or part of the Kast 
cella of the ‘alte Tempel.’ To this we 
return when the frieze is reached. 

(2) The Hast pediment.—For the centre of 
the composition Sauer’s restoration is adopt- 
ed. There was possibly not time to incor- 
porate in the translation, even as an adden- 
dum, any mention of the altered arrangement 
by Mr. Six which appeared in the Jahrbuch 
this summer (1894, 2, p. 83). For the 
remaining figures Dr. Furtwingler adopts 
the Horae of Prof. Brunn, and restores to 


































THE CLASSICAL REVIEW. 


us the ancient Moirae. To this attribution 
the authorities of the British Museum have 
steadfastly adhered, though the faith of 
many was shaken by Dr. Waldstein’s bril- 
liant theory of Gaia and Thalassa. The 
Gaia and Thalassa attribution never, I 
believe, made much way in Germany; but 
it had one great merit which should never 
be overlooked, and that is that it lent to 
the pose of the semi-recumbent sister a 
beautiful and fitting motive. So fitting do 
I feel this to be that I would suggest 
that Dr. Waldstein’s Gaia and Thalassa at- 
tribution may be combined with that of the 
Moirae. Moirae certainly the three figures 
are, but may there not have been a sort of 
underthought that the Moirae ruled over 
the three departments of the universe, 
Heaven (Ourania was the eldest of the 
Moirae), Earth, and Sea? We should thus 
gain a further point and be able again to 
compare the pediment with the Homeric 
Hymn to Athene— 


péyas 0 edeAiLer’ "Ov pros 
dewov bd Bpiyn yAavxwridos: dpdi 8 yata 
opepdadéov idxnoev: exw7yOn 8 dpa 7 b6vto0s 
KUpac Tophupeoiat KUKWULEVOS. 


And the third Fate, she of the sea, would 
recline with fresh fitness close to Nux, her 
mother, as she descends into the waves. 

I throw out this suggestion for what it is 
worth, as emphasizing the cosmic note of 
the East pediment. But the Moirae were 
local goddesses at Athens, with a local cult 
carried on by the Praxiergidae, the priest- 
esses of Athene, as Dr. Furtwiingler has 
well pointed out, and this long before they 
were Olympianized. And here I hope to 
add a fresh link in their close connection 
with Athene Aglauros. 

Athene Aglauros was, as we have seen, 
herself the olive-goddess. The Fates of 
Athens were, I believe, at one period of 
their development olive-goddesses, Moipat 
Mopia. 

I would guard against misapprehension. 
I do not say that the idea of fate arose at 
Athens or in connection with the olive-tree ; 
the conception of aica, of Themis, of dvay«n 
may, nay must, have existed prior to the in- 
troduction of the olive-tree to Athens. Nor 


do I say that popia is derived from Moirae ; 


both probably sprung from a common root 
meaning division, partition, or a//otment. 
What I am convinced of is that at Athens, 
as Athene herself at one period of her 


_ development was ’A@nvais, the sacred olive- 
tree and the Mopia was the opyxds of the 


goddess, sO the Moirae were at one time 
multiple forms of that olive-tree, 


a : ¢. popias 
The scholiast on 


Aristophanes Nuhes 
1005 makes this, | think, clear enough 
kupios popia Aéyerat ¥) lepa édala Tis Oeov, ie 
the "AOnvals -~Athene Aglauros herself 
Against this, olive as against the goddess 
herself, Poseidon sent his son Halirrhothios, 
nTTnGeis THs "AOnvas éb llowadéaw dri TH TVs 
éAalas émidelSe éreuwe tov view atroo ris 
‘Adippobtov tavryy repotvra ; and the scholiast 
goes on to tell how, the missing tree, Halir 
rhothios wounded himself. As has been 
pointed out with reference to Erysichthon, 
the tree in one version becomes humanized 
as Alkippe daughter of Aglauros. Another 
version makes him attack the olive-trees 
collectively : ai lepai éAaiat ris "AGnvas dv T 
axpoToer popiac éxadoivro’ A€yovet yap on 
“Adppobvos, 6 tais IlowedaGvos, ner noe exxowat 
avrds x.7.A.! 

It scarcely needs to be pointed out that 
the olive of Athens was the fate-tree of the 
state. The city, ruined by the Persians, 
could only revive when it sent forth a new 
shoot. It was a miraculous growth, dyeipw- 
Tov, avtoro.wv, a PoBynya even to the foreign 
foe ; moreover it was za:dorpédov, nurturer 
of the sons of the land. Hesychius tells us it 
was the custom when a male child, a Branch 
of the house of Athens, was born, orépavoy 
eAalas TiHévar zpd Tov Ovpov; at the birth of 
his child Erichthonios, old Kekrops holds the 
olive-spray in his hands. 

But the fate-tree is for death as well as 
life. This comes out curiously in the myth 
of Meleager. As long as the life-brand is 
unconsumed, the brand brought by the 
Moirae, Meleager lives; with it he dies, 
According to a less familiar, and for my 
purpose very pertinent version, given by 
Tzetzes ad Lyk. 492, the fatal branch was 
pvAXas éAaias ov das, Which Althaea swallowed 
when she became with child and which she 
brought to birth. Anyhow the life of the 
one and the other were intimately bound 
up; as the Moirae at the hero's birth 


Tempora dixerunt eadem lignoque tibique, 
Ovid, Met. viii, 454. 


Abundant evidence as to the various cus 
toms and traditions about ‘life-trees’ has 
been collected by Mannhardt ( Banmhultus, 
p. 45), and it would be superfluous to enlarge 
on the matter here. The custom survives 
with us still in our Christmas-trees and in 

1 The identity of Athene and the olive-tree has 
been long ago pointed out by Boetticher, Bauwkultus, 
. 108, but he did not extend his argument to the 
Moriae, 


90 THE CLASSICAL REVIEW. 


the pleasant custom of planting a tree at the 
birth of a child. 

Another point must, however, be empha- 
sized. The image of the goddess was made 
of her olive-tree, as were those of Damia 
and Auxesia. But this is a second step on 
from the time when the goddess was the 
tree, dwelt in the tree, her life and that of 
the people intimately bound up, practically 
identical with it.! A sort of midway stage 
in religious conception is seen in the familiar 
coin of Myrrha, where the statue of a god- 
dess actually emerges from the tree itself, 
and the local Halirrhothioi attacking the 
tree and goddess with axes are driven off by 
her guardian snakes. At Myrrha, too, the 
snake, as at Athens, guarded the life of the 
state, 

To return to the pediment. It has been 
often pointed out that in the West pediment 
the spectators are the dwellers in the land, 
the local heroes of the East pediment ; it is 


1 Mr. Marindin points out to me a curious passage 
in the Homeric Hymn to Demeter (ll. 22-3) which 
seems to bear on the human olive-goddesses. Perse- 
phone shrieked aloud as Hades bore her away, but 


> Ld > / > \ = 5 uA 
ovdé Tis aBavaTw@y ovTe OynTav avOpaTav 
Heovstev pws ovd ayAadKapTor eAata. 


The éAata: has been marked corrupt and _ freely 
emended. Gemoll and Prof. Tyrrell both hold that 
it is absurd to say that Persephone was not ‘ heard 
by gods or men or olive-trees.’ But surely it is the 
best of sense.’ It is the triple division of the hearing 
universe, gods, demi-gods, mortals. It is curious to 
find that von Wilamowitz-Moellendorf Aws Kyda- 
then, p. 125, rightly refuses to emend, but yet I think 
misses the point: ‘Als einen recht unattischen Zug 
betrachte ich die ’EAata: ayAadkapra:...freilich sind es 
Nymphen die, als die niichsten dazu das was auf der 
Flur vorgeht zu horen hier ihren Platz haben, aber es 
schickt sich nicht zu emendiren : wo moglich Sampf- 
nymphen (ayAadxapmot!) hineinzubringen. Sind etwa 
bloss Spves belebt ? nur allerdings attisch ist es nicht 
die Olbiiume zu beseelen und attisch ist es nicht, den 
Olbau der Demeter zuzuweisen.’ From an authority 
at once so profound and so brilliant I venture, with 
the utmost diffidence, to differ ; but I hold the giving 
of souls to the olive-trees to be anything rather than 
a ‘recht unattischer Zug.’ 

From the Demeter Hymn I was led to examine the 
others, and I findin the Hymn to Aphrodite a passage 
that precisely explains the midway position of the 
ayAadKapmot eAatat. The son of Anchises is to be 
nurtured by the mountain nymphs (line 260)— 


ao 6? 


al pf obre Ovntots ov7 ABavdrooww EmovTa.. 


Next Aphrodite goes on (1. 267) to tell of the trees, 
the éAdra: and dpves, who also will nurse the child— 


Kadal, THAEOdovoeM, ev otpeow bWnAoio 
[éorao’ nAlBaror Tewevn 5é E KLKAHOKOVoOW 
abavatwv: Tas 8 ots Bpotol Kelpovor cidhpy] 
GAN OTE kev OH moipa TapectHKy Oavdro.o, 
a¢averat wev mp@Tov em xOovl dévdpea Kara 
paoids & aupimepipOivv0er, wimrovar 8 am Cot 
Tay 5é 8 duod WuxXh Aelret Gdos HeAtouo, 
ai mev eudy Opepovor mapa opiow vidy Exovea, 
The passage speaks for itself, 


laid down that they must be dwellers in 
Olympos. This is true, but with a difference, 
or rather an added condition. I believe that 
all the spectators of the birth of Athene 
are, asin the West pediment, local divinities, 
but such as have been Olympianized, turned 
into cosmic potencies by advanced Athenian, 
Pan-Hellenic theology. This is true of 
Helios, who had his local Eiresione with the 
Horae at Athens ; doubly true of the Horae 
who are themselves Agraulides, two-fold 
forms of Athene, before they become gate- 
keepers in the Pan-Hellenic Olympos. At 
the Horae I must pause for a moment to 
note how admirably they balance in idea the 
Moirae ; their very name has the same fun- 
damental conception ; they are the division, 
the allotment, partition of time, whether of 
day, month, or year. As the Moirae of life 
they gradually absorb to themselves the 
notion of the right time, the prime, beauty, 
and vigour. Hence, as Dr. Furtwiingler 
beautifully points out, they are on the side 
of Helios, the Moirae on the darker side of 
Nux, the Horae of happier nature, the 
Moirae of sadder humanity ; but both are 
Aglaurides, both manifestations, double or 
triple forms of Athene. I hesitated defi- 
nitely to call the Moirae Aglaurides tree- 
goddesses, though I knew them to be so, but 
Hesychius is bolder: ’AyAavpides, potpar 
mapa A@nvaiors. Of course the simple state 
ment has been emended: ‘forte legendum 
vipat tapa "AOnvaiors, forte teper.’ Any- 
thing rather than the simple truth. 

The Moirae, then, I hope bear out my 
principle that the spectators present at the 
birth are local deities Olympianized, become 
cosmic potencies. So reinforced I turn to 
the so-called ‘Theseus,’ who cannot, to my 
mind, be Kephalos. Kephalos in the key of 
the pediment, as I understand it, strikes a 
jarring false note. Heis Attic enough, but he 
never became Olympian, never in all the 
hey-day of his popularity cosmic or even Pan- 
Hellenic. Dr. Furtwiingler is perfectly right 
when he says, turned as the figure is south- 
wards it must be connected with Helios ; 
so our choice is small. There is one god, 
and one only, who fulfils all the conditions 
Dr. Furtwingler justly lays down, and 
mine also, that he should be Pan-Hel- 
lenic and cosmic ; one god who is a beautiful 
youth, a mountain-god, a hunter, who 
watches for the rising of the sun; who, 
moreover, is closely allied to the Horae, to 
whose dancing he pipes. That god is Pan. 
On his close connection with the sun in 
Arcadian cults I need not here enlarge ; the 
subject has been fully dealt with by Immer- 






















THE CLASSICAL REVIEW, 


-wahr, and more recently by Berard, both 


of whom conclude that he was a primitive 
sun-god. ‘Two points I may add. On a 
curious suprise vase figured in Gerhard, Licht- 
gottheiten (Taf. iii. 3), Pan, a torch in his hand, 
actually leads by the rein one of the sun- 
god’s horses as he mounts from his boat in 
the sea. I cannot of course lay very much 
stress on a monument of late date, and so 
far as I know unique, especially as the 
figure on the Blacas sunrise vase in the 
British Museum (op. cit. i. 2) which has 
usually been called Pan, must now, as Mr. 
Cecil Smith kindly informs me, in all prob- 
ability bear another name. The second 
point is that it cannot be forgotten that 
Pan-worship, if not actually introduced, was 
at least revived in Athens after the battle 
of Marathon ; and it is in the fifth century 
that the Agraulides begin to figure in art 
and literature as dancing in the caves that 
were Iavos Gaxjpata. There is one objection 
that many will deem fatal to my theory, 
and in stating it I cannot alas! take the 
god by the horns. In types of Arcadian coins 
of the fifth and fourth centuries z.c, Pan, 
even when represented as fully human and 
seated, as he often is, on a rock, much after 
the fashion of the ‘ Theseus,’ has uniformly 
the adjunct of horns. I have carefully 
examined the forehead and hair of the 
‘Theseus’ and can find no trace whatever 
of anything of the sort. I do not believe, 
however, that they are essential to the god. 
He appears in endless variety of transition 
stages from goat to man, and his identity 
could be placed beyond a doubt without 
horns by the attributes of syrinx and pedum. 
In the Homeric hymn, when Hermes takes 
his baby-boy to Olympos, the whole point of 
the scene is that the shaggy prick-eared 
thing should move the blessed gods to 
laughter; but in the grave pediment of 
Pheidias the solemn satellite of the sun had 
other functions. 

(3) The Central Group of the East Frieze. 
—‘Dass in der Mitte der Ostseite dieses 
Frieses die Uebergabe des Peplos dargestellt 
ist, hiitte man nie bezweifeln sollen.’ The 
dogmatism of the ‘sollen’ stirs in us the 
demon of flat contradiction. It is, I hold, 
the sacred duty of every intelligent archaeo- 
logist to doubt the presence of the peplos. 
We are weary of hearing that the slab with 
the presentation of the peplos occupies the 
central place in the frieze ; it does no such 
thing. The centre point is the figure of 
Athene’s priestess, because she is the repre- 
sentative of the goddess ; the so-called * pre- 
sentation of the peplos’ is well to the right, 


balancing of course the diphrophoroi, Ag 
is it likely, is it reasonable, to SUP Pose 
the peplos, worked with solemn care by 
maidens whose life was dedicated to th: 
goddess, who dwelt apart under the cave of 
her priestess, should be offered by m mere 
lad, and offered not to the priestess but toa 
priest? Maidens are thronging in the pro 
cession carrying sacred vessels: was there 
never a maiden ready to make the supreme 
offering And last, whatever the necessities 
of frieze perspective and composition, how- 
ever much the scene is conceived, as it 
assuredly is, as taking place within the 
temple, would any artist in his senses have 
so arranged the slabs that Athene should 
actually turn her back on the gift offered 
to her ¢ 

One signal service, however, to the right 


interpretation Dr. Furtwiingler has done. 


His explanation of the diphrophoroi carries 
immediate conviction—they are preparing 
the seats for the theoxenia of the twelve 
Olympians actually seated outside. Dr. 
Furtwiingler tries, but in vain, to make out 
a balance between preparations for a banquet 
and the offering of the peplos. Quite in 
vain—the one conception is Olympian, Pan- 
Hellenic ; the other most strictly local. It 
was reserved for Prof. Curtius, as the crown 
of his old age, to see the truth by the light of 
the Magnesia inscription (Anzeiger, 1894), 
and to deal what is, we trust, the final death- 
blow to the ‘peplos theory.’ The huge, 
heavy, stiff object brought by the boy, 
received by the priest, is a orpwuvy, a carpet 
to be spread when the banquet is set. 
Athene may turn her back as she will, it 
is for her servants, and most fitly a man 
and a boy, to attend to the furniture of the 
feast. In the inscription full instructions 
are given that the twelve gods, their actual 
xoana, should be brought out in their festal 
raiment. A. tholos is erected to shield them, 
and the order is given orpewveebar orpepras. 
Curiously enough Dr, Furtwiingler, in the 
new edition, cites the inscription as support- 
ing his view of the diphrophoroi, but does 
not take the natural second step on to the 
otpwpvy. It is possible, as there is no 
mention of Prof. Curtius’ theory, that it 
was not known in time even for an adden- 
dum ; if so, it will be interesting to note if 
Dr. Furtwiingler agrees. 

The offering of the peplos, as every one 
agrees, was addressed to the old olive-wood 
xoanon, not to the ready draped chrysele- 
phantine statue of the Parthenon. The 
best proof, Dr. Furtwiingler says, that the 
intention was to transplant the old xoanon 


92 THE CLASSICAL REVIEW. 


with its cult to the Parthenon, is the scene 
of presentation on the frieze. The image, of 
course he knows, never was moved to the 
Parthenon ; it went to the East cella of the 
Erechtheion, the temple év @ 7d dpxatov 
ayaApa. Here he is at issue with Dr. Dorp- 
feld, who, holding still to his view that the 
‘alte Tempel’ was standing in the days of 
Pausanias, believes that the xoanon never 
left this building, never passed into the 
Erechtheion at all.!. With it remained the 
sacred lamp of Kallimachos, in which sur- 
vived the hearth-fire of the city. Kalli- 
machos therefore need not shift his date, 
and on this view the ingenious structure of 
Dr. Furtwiingler respecting the archaic 
artist falls to the ground. On this matter, 
the date of Kallimachos—interesting to the 
student of sculpture rather than to the 
mythologist —I will not dwell; but the 
question of where the xoanon finally re- 
mained is all-important in relation to Pau- 
sanias as well as to mythology, and I cannot 
pass it by. In my book on Athens (M/yth- 
ology and Monuments, p. 508) I stated that, 
when Pausanias writes xetrar d€ év TO vad 
THs LoXduddos, he refers, not to any shrine or 
division of the Erechtheion, but to the 
‘alte Tempel,’ the contents of which he 
proceeds to describe. When he describes 
the Erechtheion he ealls it the Erechtheion, 
oiknpa ’EpexGetov ; in this matter, and this 
only, I ventured to differ from Dr. Dirpfeld. 
I did not, however, venture to place in the 
‘alte Tempel,’ as Dr. Diérpfeld now does, the 
xoanon and the Kallimachos lamp, because 
of the Chandler inscription, Pausanias gave 
no clue to their whereabouts, as he intro- 
duces the notice of them with the vague 
exordium iepa pev tHs “AOnvas x.t.A. But 
with what joy I carry them back to the 
‘alte Tempel’ I hope to express more fully 
when Dr. Dérpfeld prints his evidence. The 
text of Pausanias now reads clean and clear. 
He describes first the Erechtheion, with the 
altars of Poseidon, Butes, and Hephaistos, 
the stranger gods, the well which Erechtheus 
gave to Poseidon when their cults were 
fused. He does not describe the Kekropion, 
for in the Erechtheion, according to my 
view, Kekrops had no part or lot. With 
the vague though imposing exordium tepa 
pev THs A@nvas x.7.X. he passes into the ‘alte 
Tempel’ and sees the old xoanon and the 

1 In a letter just received from Dr. Dorpfeld he 
kindly gives this view with some comment on the év 
@ 7d apxaiov &yadua of the Chandler inscription and 
the Strabo passage which seem to block the way, but 
as a full statement from himself may, we hope, be 
looked for before long, I will not anticipate its publi- 
cation, 


ever-burning lamp, inseparable I believe, for 
the Polias guards the hearth. In the temple 
of Polias he sees a Herm of wood, said to 
be an offering of Kekrops; and where, I 
ask, could he rightly see it but in the Kek- 
ropton ?—the Kekropion which is no part of 
the Erechtheion, though outside, close up to 
the porch of the Caryatids (rH tpoorace THe 
apos ta[t] Kexporiwr). The Herm was prob- 
ably no Hermes at all (what should Hermes 
be doing there ?), but a primitive image of 
Zeus Soter, z.e. Kekrops himself, the ancient 
husband of the olive-tree xoanon. 

_ To return for a moment to the frieze. 
The offering of the peplos was undoubtedly 
a portion of the Pan-Athenaic festival; so 
was the bringing of an Hiresione, so was the 
ddoAvypos, so the Pyrrhic dance, and no 
doubt many another primitive rite of a 


primitive harvest festival ; but it was not 


these that Pheidias cared to embody in his 
Pan-Hellenic temple. Athene is no longer 
the primitive mother-goddess of a simple 
agricultural people ; she no longer needs to 
renew her virginity year by year with the 
putting on of a new robe. Athene Parthenos 
cared no more about her new peplos than 
she did about the yearly bath essential to 
the primitive xoanon at the Plynteria. It 
is not really the Pan-Athenaic festival in its 
entirety that Pheidias represented at all; 
he did not even choose out elements ‘ suited 
for artistic representation’ or reject elements 
‘plastically unpleasing.’ In the frieze he 
frankly takes the theoxenia offered to the 
twelve gods by Athene on her birthday 
festival. It is the Olympian, Pan-Hellenic 
feast that is the gist of it all. The real 
apex and culmination of the procession is 
the group of the twelve gods; it is all- 
Athens entertaining all-Olympus. Had our 
minds not been obscured by the peplos 
theory this must have been obvious long 
ago; it is the triumph of the canonical 
Olympian, Pan-Hellenic faith over the local 
cult, and to obtrude the peplos is to make a 
fiasco of the whole. In the interpretation 
of the frieze, between Dr. Furtwangler and 
Prof. Curtius the honours are divided. 

I have been concerned to criticize, and 
have approached a book, mainly artistic, 
from a purely mythological ground, If I 
have felt that the mythology of the author 
has its weak places, I have also felt wonder 
and admiration that, where mythology is 
merely a parergon, it should yet be consis- 
tent though incomplete. If I have criticized 
here and there I have learnt almost every- 
where, 

JANE E, Harrison, 

















































THE CLASSICAL REVIEW 


MONTHLY RECORD. 


BRITAIN. 

Darenth.—A Roman villa has been found, the 
foundations being quadrangular, and the outer walls 
made of flint nodules set in mortar and faced with 
plaster. The floors are paved with small cubes of red 
brick, or with square red tiles. Very few small anti- 
quities have been found—some fragments of bronze 
armlets, chains, a so-called hippo-sandal, a coin of 
Tetricus, &c.—but broken pottery is abundant.! 

Great Chesterford.—Prof. Hughes has discovered 
near the Roman Camp remains of Samian and other 
pottery. All the remains in this locality point to 
the existence of a permanent Roman town rather 
than a temporary military station.* 


SPAIN. 

Tarraco.—A Roman bronze bell has been found 
with an inscription of the end of the second century, 
containing the word cacabulus, a new term for ‘ bell.’ 
The bell served sacris Atugustis.* 


ITALY. 

Pompett.—At Boscoreale, Signor Prisco is exeavat 
ing a large house, the bath-room of which seems to 
have been dug out some time ago and its contents 
removed. ‘He has found two cisterns for supplying 
the bath and basins with hot and cold water. These 
cisterns stand at the side of a large atrium, probably 
used as a kitchen, with the hearth-place in the 
middle. On another side are traces of a wooden 
cupboard or sideboard. On the hearth remained the 
cinders of a fire. In one wall is a niche for the lares 
and penates. The bath-rooms consist of an ante- 
chamber, with two ducks represented in mosaic on 
the floor ; the éepidariwm, with the figure of a large 
fish; and the caldariwm, with a swan or crane 
stretching out a claw towards an ee]. All the metal 
pipes, taps, &c., still remain in the bath-room. The 
bath is marble-lined and of the usual- size for one 
person. A niche in the form of a shell doubtless 
contained the basin. In the heating-room behind 
the bath is a leaden boiler more than six feet high 
and two feet in diameter. An elaborate system of 
pipes connects the water-cistern with the boiler and 
other parts of the bath. Many domestic utensils, 
seals, &c., have been found on the spot.# 


GREECE. 
Attica, Aphidnae.—A tumulus containing twelve 
pe of Mycenaean character has been excavated. 
n the graves were charred skeletons, one of colossal 
proportions. Various metal ornaments were found at 


the same time.® 


Prasiae. —The prehistoric necropolis is being 
excavated. More than 200 vases (Mycenaean in 
form, but with some hitherto unknown decorative 
designs), two sword-blades, and three rings, one of 
gold and two of silver, have been found. 

Epidaurus.—The base of a statue has been dis- 
covered with the name of Thrasymedes of Paros, 
the contemporary of Praxiteles.” 

Delphi.—The frieze of the Treasury of Siphnos 
and the metopes of that of the Athenians, like those 
of the Treasury of the Sikyonians, bear painted 
inscriptions in red or black, explaining the figures. 





1 Times Weekly, 7 December. 

2 Academy, 15 December. 

3 Berl. Phil. Woch. 24 November. 
4 Athenaeum, 22 December. 

5 Berl. Phil. Woch. 15 December. 
® Athenacum, 8 December. 

7 Athenaeum, 29 December. 


mn , 
Phe subjects and the personages, so far as the 
have been deciphered, are as follows :-— 7 


Sikyon. The rape of Europa The ram arTy ing 
Helle. The Calydonian boat The Messenian 3 tv 
dition of Castor, Pollux and Idas The ship Argo 
Treasury of Siphnos, Apotheosis of Herakle 


Gigantomachia. Combat over the body of Sarpedon 
and the Gods looking ou from Olympus. Contest of 
Pelops and Oenomaus. Treasury ad Athens Gi 
gantomachia. Labonrs of Herakles and of Theseus." 

Ainorgos. Tsountas has discovered, in the oour™ 
of his excavation of a prehistoric necropolis, some 
twenty tholos-tombs, containing terra-cotta vases. 
lance-heads, and statuettes (one of marble). Theis 
age is supposed to be not more recent than the ee ond 
millennium nc." 

ASIA MINOR. 

M. Ernest Chantre has reported to the Académie 
des Inscriptions on the archaeological mission to Asia 
Minor on which he was sent by the Minister of 
Public Instruction. He found cuneiform inscriptions 
at Boghaz-Keui (Pterium) and at Kara-Euyuk. near 
Caesarea, where is the site of a ‘ Pelasgic' city. His 
discoveries extend further west the known ‘area of 
Assyrian influence, and may throw light on the 
sources in Asia Minor of ‘ Mycenaean’ civilization.” 

AFRICA. 

Alexandria,—Botti is-said to have made interest- 
ing finds near Pompey’s Pillar, on the supposed site 
of the citadel of Alexandria." 

Carthage.—Pere Delattre has excavated two new 
tombs in the Punic necropolis. In one, of rectangu 
lar form, dating from the sixth century, was a 
skeleton of a man of Phoenician type, with rich 
objects around him. In the other were found a vase 
of fine black clay, a goblet of red clay with black 
iine ornamentation, an incense-burner of brown clay, 
a Punic lamp, some shells, a bronze axe, a mirror, 
two alabaster vases, scarabaei, statuettes of Anubis 
and Ptah, and ornaments in silver and agate,* 

G. F. Hint. 
SUMMARIES OF PERIODICALS. 

Archdologisches Jahrbuch. 1894, part 1. 

1. Pallat: publishes (pll. 1-7) the remains of 
sculptures found by Staés in 1890 on the site of the 
temple of Nemesis at Rhamnus : they belong to the 
reliefs on the basis of the Nemesis statue (Pans. i 
33, 7) and are probably from the hand of Pheidias : 
on p. 9 a suggested restoration is given. 2. von 
Gaertringen : examines the inscriptions which bear 
upon the Rhodian school of artists, and concludes 
that the activity of these artists must have covered 
two centuries: the later series, comprising the larger 
number, must have worked in the first half of the 
first century B.C. : the flourishing periods of art cor- 
respond with the political history of Rhodes, and 
terminate in B.c. 43. 3. Firster: publishes a on 
ment of a lamp and an impression of a gem, both 
representing the Laocoin groups the gem was form. 
erly in the possession of the prior of Tyrwardireth in 
Cornwall (1507-1539) but has now disappeared : it is 
uncertain whether it is antique ; its subject differs in 
several details from that of the Vatican group. 4. 
Studniczka: argues that, in Pausanias’ deseription 
of the chest of Kypselos (v. 17, 9), there is really no 
difficulty in assigning the seated Heracles to the 
scene of the funeral games of Pelias : Heracles here 


8 Comptes R. de U Acad. des Inser, October, 
® Athenacum, 24 November. 

1” Academy, 17 November. 

1 Academy, 8 December. 

12 Athenaeum, 5 January. 


94 THE CLASSICAL REVIEW. 


performs the same function as that which Homer in 
the funeral games of Patroclos assigns to Phoenix, 
that of marking the répua. 5. Hauser : republishes 
a cut of the Mycenaean relief in the British Museum 
(Cat. Sculpture i. no. 5): it is not a lion but a bull 
charging : its provenance from Mycenae is probable, 
and thus it is an additional argumeut in favour of 
the Vaphio cups being local products. 


Anzeiger.—Notice of Lolling. <A detailed notice 
(by F. Winter, pp. 1-23) of the Sidon sarcophagi, 
with cuts. Acquisitions, Dresden 1892. Meetings 
of the Arch. Gesellschaft at Berlin: (papers by 
Curtius on a painted pinax, Berlin Cat. no. 2759: v. 
Rohden on the so-called ‘Campana’ terracotta 
reliefs : v. Gaertringen on the history of the theatre 
at Magnesia : Curtius on ‘ the Achaeans at Olympia’ : 
Winter on a female statue from Pergamon: and 
Briickner on the year’s excavations at Hissarlik). 


The Same. 1894, part 2. 

1. Milchhofer: publishes an ‘aryballos’ in the 
Louvre, and discusses the whole subject of the later 
Attic vases: on pp. 58-63 he gives a list of forty- 
four typical examples, divided into an earlier and 
later series, and grouped chronologically: he con- 
siders, from historical and internal evidence, that 
Attic vase-painting must have passed through all the 
phases of its development known to us before the 
end of the fifth century. 2. Six : proposes a restor- 
ation of the central group of the east pediment of 
the Parthenon, based on the Madrid Puteal and 
Sauer’s plan: Zeus is seated on a throne to nr, 
Hephaistos and Athene (of smaller size) move away 
on either side, and Nike, attached to the tympanum, 
crowns Athene. 3. Wernicke: notes on Olympia : 
(1) its altars : shows that there is no reason to consider 
the account of Pausanias as defective or patchwork ; 
and explains the topographer’s method: (ii) the 
history of the Heraion : shows that the chief struc- 
tural changes were a consequence of the visit of the 
Emperor Nero. 4. Stengel: disputes Mayer’s inter- 
pretation (ante, viii. p. 218) of the word orAdyxva. 

Anzeiger.—Annual reports of the Institute. 
Meetings of the Arch. Gesellschaft: (papers by 
Winnetield on the Villa of Hadrian at Tivoli: by 
von Fritze, on a fragment of an alabaster cup from 
Naukratis (B.4f. Cat. of Sculpture, no. 116): Weil, 
on the numismatic evidence of the ’A@. ToAtre:a : 
Kekulé on the Magnesia excavations: Kern, on the 
temple of Zeus Sosipolis at Magnesia: and Heyne, 
on the Artemision found there: Adler, on the altar 
of Zeus at Olympia : Briickner, on the development 
of Trojan pottery : Winter, on a marble portrait head 
in the Louvre, representing Mithradates VI. Eu- 
pator). 


The Same. 1894, part 3. 

1. Graef: discusses the heads of the Florentine 
group of wrestlers : suggests that the heads belong 
to the group, but have changed places: the group is 
a work of the fourth century, non-Attic, but free 
from Lysippian influence, dependent on the tradi- 
tions of Scopaic art. 2. Wernicke : notes on Olym- 
pia, continued : (iii) the Proedria and the Hellano- 
dikeon. 3. Briining: on the prototypes in art of 
the tabulae Iliacae: both the Ilias Latina and the 
tabulae are traceable to the influence of a large 
number of works of art. 4. Klein: in reference to 
Mayer's article (Ath. Mitth. 1892, p. 261), points out 
that there is yet a third group of Thespiadae, 
referred to by Pliny : the word thespiadas in omitted 
in the Bamberg MS., but given in the others, and is 
for other reasons probably the correct reading. 

Anzeiger.—Notice of Brunn. Von Duhn describes 
an oil-painting in the entrance hall of the University 


at Amsterdam, which he shows is the earliest known 
view of Palmyra: photographs of it can be obtained 
from the Arch. Inst. at Berlin. Acquisitions, Berlin 
Antiquarium. Meetings of the Arch. Gesellschaft : 
(papers by Kern on Aitemis Leukophryene, giving 
the text of an inscription (first half of first century 
b.C.) which refers to the setting up of a xoanon of 
Artemis eis Tov kareockevacpevoy ath viv Mapdevava : 
von Gaertringen on a dedication by an astronomer, 
found near Lindos. Notice by Reisch of the meeting 
of anthropologists at Innsbruck and of the exhibition, 
arranged for the meeting, of the Lipperheide collection 
of ancient bronzes). 

"Epnuepls "Apxaodoyinh. 1893, part 4. 

1. Nicolaides : disputes the identification of His- 
sarlik as the site of the Homeric Troy: neither the 
literary references nor the topographical features suit 
this site, but suit: Bunarbashi better. He also quotes 
and comments on the literary evidence which he 
thinks supports the old view, that the Enneakrounos 
is the Callirrhoe of the Ilissos, where the excavations 
of the Archaeological Society have recently laid bare 
remains of architecture and reliefs of a fine style. 
He thinks the ‘Odeion’ mentioned by Pausanias in 
this connection is uncertain by reason of the bad 
condition of the MS. here: but he does not attempt 
to explain Pausanias’ abrupt change in his route. 
2. Cavvadias : publishes (pll. 12-138) a bronze figure 
of Zeus Ammon with rams’ horns and body termin- 
ating in the forepart of a serpent: probably of 
Alexandrine period. 8. Mayer: publishes (P]. 14) 
fragments of a pedimental group from Eleusis repre- 
senting Pluto carrying off Persephone in the presence 
of other deities. 4. Leper: fragment of a catalogue 
of prytaneis, of about 408 Bc. 5. Mylonas: 
publishes (plate 15) a bronze folding mirror from 
Eretria : on one side it has a relief of a woman feed- 
ing a swan upon which she rides, probably Aphro- 
dite : on the other a woman on a horse springing up 
from the waves, probably Selene. 6. The same: 
various sepulchral inscriptions. 


The same. 1894, parts 1 and 2. 

1. Dorpfeld : an answer to Nicolaides’ article noted 
above: he thinks that when Nicolaides has again 
visited Hissarlik and sees the magnificent remains of 
the Mycenaean period recently laid bare there, and 
compares them with the remains of Bunarbashi, he 
will be converted. As to the Enneakrounos, the 
literary passages only go to strengthen his opinion 
that the Enneakrounos was near the Agora, an 
opinion confirmed by the German discovery of natural 
springs and a great Hellenic conduit at this very 
spot. 2. Cavvadias: publishes (pl. 1) two reliefs 
found in the excavations of the Asklepieion at Epi- 
dauros, and which he considers are copied from the 
chryselephantine statue of Asklepios made by Tbrasy- 
medes. 3. The same: inscriptions from Epidauros. 
4. White: discusses the signification of the term 
Pelargikon in the time of Perikles, in view of the 
various literary references. He concludes that, at 
that period at least, the Pelargikon as a fortification 
did not exist, and that the Acropolis was not a 
stronghold, and was not regarded as such by the 
Athenians. 5. Staés: publishes (pl. 2) a white 
lekythos from Eretria with a somewhat unusual ver- 
sion of the typical mourning scene : he believes it to 
be of Eretrian fabric. 6. Nicolaides: in publishing 
(pl. 3) three works of art giving reminiscences of the 
Iliad, expresses generai views on the subject: he 
thinks that ‘apart from allegory and the relations ot 
gods and mortals, the descriptions in the Iliad of places, 
men and actions have nothing fanciful or mythical’ ; 
and returns to the.attack on Hissarlik as the site : 
with an explanation of the famous silver relief from 





















































a i le ee 


THE CLASSICAL REVIEW. 


Mycenae. 7. Philadelpheus : discusses the type of 
the Gorgoneion in the centre of the Peiraeus mosaic 
in connection with other types of this subject. 8. 
Millet: publishes (pl. 5) a mosaic from the church 
in Daphnion represeuting the crucilixion, a work of 
the first half of the eleventh century. 9. Hartwig: 
publishes (pl. 6) a vase in form of a negro’s head, 
with Leagros kalos. 10. Skias: once more as to 
the reading cagr%p (ante 1892, p. 256). Obituary 
notice of Lolling, by Mylonas. : 

The same. 1894, part 3. 

1, Skias: publishes (pll. 7-8, and eut) four reliefs 
found in excavations in the bed of the Ilissus. 2. 
Homolle : by comparison with three other inserip- 
tions of which he gives the text, assigns the testa- 
ment of Epicteta to about b.c. 210-195. 3. Millet: 
the Daphnion mosaics, continued ; the birth of John 
the Baptist (pl. 9). 4. Skias : a series of inscriptions 
from Eleusis: among them is part of an artist’s 
signature ending...Aivato: éménaa(y). 5. Mylonas: 
publishes (pll. 10-11) six fragments belonging to four 
metopes from the south side of the Parthenon, 
showing their position as determined by Carrey’s 


drawings. 


Athenische Mittheilungen. 1894, part 3. 

1, Preger and Noack : an account of the results of 
some researches made in 1893 on the site of Dory- 
laion : a sketch map of the site, a series of inserip- 
tions, and a series of cuts of tombstones on which 
are sculptured in compartments various objects and 
utensils of daily use: remarks on the historical 
development of this custom. 2. Six: explains the 
disputed Eriphyle motive (Paus. x. 29, 7) by com- 
parison with a bronze statuette in Athens (cut). 
3. The same: publishes a limestone inscription in 
the Museum at Corfu, recording that the stone (of 
conical form) was set up by Mys: it represents an 
Agyieus, and may be compared with two similar 
stones at Pompeii. 4. Korte: a Boeotian vase with 
a burlesque scene: two comic figures, engaged in 
braying objects in 2 mortar, drive away two geese : 
though represented in the costume of Phlyakes, they 
are not actually on the stage: as in Italy, so in 
Boeotia, such figures are borrowed originally from 
the Attic stage, and reappear in untheatrical scenes, 
merely as burlesque clowns. This accounts for the 
burlesque representations of myths on the Cabirion 
vases. 5. Pernice: (i) describes the finding of one 
of the inscribed*boundary stones which marked the 
division between Messenia and Lacedaemon, about 
one and a half hours east of Sitsova: another is 
known to exist between it and Chani: (ii) proposes 
to identify Janitsa with the site of the ancient 
Pherae, and publishes two inscriptions from a chapel 
near: (iii) identifies the road over Taygetos by 
which Telemachos went by chariot from Pylos by 


_ Pherae to Sparta. 6. Forster: publishes five in- 


scriptions from Bithynia ; three are metrical epitaphs, 
the others are dedications to Zeus ’Emdjusos and 
Zeus BaAnos, both titles otherwise unknown.' 7. 
Dragoumes: examines the passages bearing upon the 
*AuaCovls or7jAn (the monument of Antiope) and 
proposes in Paus. ii. 1 to amend émel éa7A@ov (sc. 
eis thy méAwv). 8. Dorpfeld: an account of the 
excavations at Troy in 1894: the chief object was to 
completely uncover the acropolis of the sixth stra- 
tum: pl. 9 gives the results, showing almost 





1 Among the inscriptions copied in N. Phrygia by 
Messrs. Munro and Anderson, and which will shortly 
be published in the J.7.., is one recently dug up 


near Sinekler, recording a dedication to Zeus Pande- 
mos: this epithet, equally unknown hitherto, scems 
to bear comparison with the Bithynia inscription. 


95 
exclusively the buildings of this stratum om 
oes the Pelasgic walls still remain several metres 
igh: three towers and a door were found. The 
results will be more fully published later, 9, Fras 
kel : answers Wilhelm \anle, p. 294) as to the Hix 
pomedon inscription. 10. Kern: publishes a new 
list of Theori from Samothrace. 11, Pollak « pub- 
lishes an inscription from Athens of the second 
century A.D., apparently a list of names 

Revue Archéologique. July—August. 1894. 

1. Deloche : list of Merovingian seals and rings, 
continued. 2. Le Blant: discusses the old super- 
stition as to the prophylactic qualities of the first 
chapter of S. John and quoting instances of its use, 
3. Torr: publishes fifteen new representations of 
ships on Dipylon pottery (cuts) with comments on 
the class of ships indicated : on p. 16 note 1 is given 
a list of those previonsly published. 4. Legrand : 
notes on the Parthenon marbles (cf. ante, i. p. 76) 
based on an examination of the papers of Fauvel 
bearing on the question as to whether Gaspari or 
Fauvel really acquired the Parthenon marbles for 
the Louvre. 5. Nicole: a petition addressed to a 
centurion by Egyptian farmers : it is a papyrus, and 
bears the name of the Prefect Subatianus Aquila 
(207 A.D.). 6. Espérandieu: indices to his lists of 
Roman oculists’ stamps, continued. 

8. Reinach’s Chronique d’Orient (pp. 62-120). 
Meetings. Reviews. 

The same. September—October. 1894. 

1. Le Blant: obituary notice of de Rossi. 2. 
Publication (pll. xi.—xv.) of an ivory bex in form 
of a head, found in 1878 at Vienne (Istre) and now 
in the Museum there: with notes by Bertrand, 
Maitre and S. Reinach. 3. Espérandieu: Roman 
oculists’ stamps, concluded : chronological classifica- 
tion, supplement, bibliography, and addenda. 
4. Carton: Punic stamps on amphora handles. 
5. Daressy : the great towns of Egypt at the Coptic 
period. 6. Legrand: notes from the national 
archives bearing on the Choiseul-Goutflier collection. 
7. Magon: publishes cuts of a leaden object in the 
Marseilles Museum, which he restores as part of an 
anchor: he is apparently unacquainted with the 
anchor recently acquired by the British Museum (see 
Torr, Ancient Ships, pl. 8, figs. 45-47). 8. Brinicky: 
notes on the history of the Palatine mount: extracts 
from a treatise in Greek. Meetings: Reviews. 

C. 8. 


- it &o 


Numismatic Chronicle. Part iii. 1894. 

Arthur J. Evans. ‘Contributions to Sicilian 
Numismatics’: 1. On the recent discovery of a 
Damareteion from a new die, 2. The place of the 
Damareteion in the Syracusan series. 3. On a hoant 
of archaic and transitional Sicilian coins recently found 
at Villabate near Palermo, 4, Some new lights on 
the monetary ‘frauds of Dionysios. 5. The effects 
of the Dionysian finance on the silver systems of 
Etruria and Rome. 6. The omen of the battle of 
the Krimisos on coins of Herbessus and Morgantina. 
7. The African gold stater of Agathokles. 


Revue belge de Numismatique for the year 1804. 

Contains ‘ Britomartis la soi-disant Europe sur le 
platane de Gortyne’ by J. N. Svoronos, pp. 113 
147 ; and a note on the Empress Sulpicia Dryantilla 
and her coins by Cte. Maurin-de Nahuys, pp. 259 
285. 

Recue Numismatique. Part iv. 1894. ; 

R. Mowat: ‘ Kelaircissements sur les monnaies des 
mines.’ A study of the bronze coins of Trajan and 
Hadrian inscribed ‘ metallum ' and bearing the names 
of mines in Dalmatia, Pannonia &e. me speci- 
mens are marked 8.C., but Mowat supposes that the 


96 THE CLASSICAL REVIEW. 


majority were struck at Viminacium in Upper Moesia. 
The types of the divinities Apollo, Diana, Mars and 
Venus are ingeniously explained as respectively 
indicating mines of gold, silver, iron and copper. A 
good list of the coins is given.—Th. Reinach. ‘Un 
nouveau roi de Paphlagonie.’ On a bronze coin in 
the British Museum of Deiotarus Philadelphus show- 
ing on one side the name and effigy of King Deio- 
tarus Philadelphus, and on the other the pilei of the 
Dioscuri and the name of King Deiotarus Philopator 


Jahresberichte des Philologischen Vereins 
zu Berlin. January—May, 1894. 

ON THe LITERATURE OF CurRTIUS, by M. Schmidt. 

I. Editions. Vogel’s 2nd vol. has not yet appeared 
in a new edition, and the edition of Linsmayr has 
not been completed since his death. Three French 
editions have appeared in 1890, 1891 by A. Aderey, 
A. Vanchelle, and G. Delbes respectively, but 
nothing is said of them. 

II. Text-criticism. F. J. Drechsler (Zeitschr. f. 
dst. Gymn. 1890, p. 193) on iv. 1, 3 reads Jam ocius 
tum. P. Prohasel, Q. Curtiit Rufi codicum me- 
moriae emendandae leges et proponuntur et adhi- 
bentur. Pr. Sagan 1890. The codd. BFLV fall 
into two classes, LV and BF, of which the former is 
freer from errors. Lacunae, additions, and altera- 
tions are discussed. ; 

III. Lexicons. O. Hichert, Vollstandiges Worter- 
buch zu dem Gleschichtswerke des Q. Curtius Rufus. 
3rd edition, Hannover 1893. Much improved in 
many places. 

IV. Language. J. Steinhoff, De usu nominwin 
urbium insulurum terrarum Curtiano. Diss. Frei- 
burg 1883. OC. on the whole follows the rule about 
the employment of prepositions before proper names. 
Max C. P. Schmidt, Ac wnd atque vor Konsonanten. 
Fleckeisens Jahrb. f. Phil. 1889. CC. like Caesar 
and Livy prefers ac to atgwe before consonants. F. 
Knoke, Ueber den Gebrauch von plures bei Q. Curtius 
Rufus. N: Jahrb. f. Phil. 1891. Maintains that 
plures is always a comparative and never=complures. 


(Baothéws Anuo[tdpov birow]aropos). Philadelphus is 
mentioned by Strabo (xii. 3, 41) as being the son of 
Castor (IJ. ?) and the last king of Paphlagonia (x.c. 
32-5 ?) and is also known from a rare drachm in the 
Berlin Museum. Philopator is quite unknown, but 
Reinach shows good reason for supposing that he 
was the brother and predecessor of Philadelphus 
reigning as King of Paphlagonia (B.c. 36-32 2). 


WoW 


Weinhold, Bemerkungen zu Q. Curtius Rufus. Pr. 
Grimma 1891. Some remarks on words expressing 
number. Rauch, Gerwndiwm und Gerundivum bet 
Curtius. Pr. Meiningen 1889. <A. Ludewig, 
Quomodo Plinius Major, Seneca philosophus, Curtius 
Rufus, Quintilianus, Corn. Tacitus, Plinius Minor 
particula quidem wsi sint. Fase. I. Prager Phil. 
Stud. 1891. 

V. Dissertations. I. Evers’ Kritik des Fraen- 
kelschen Buches. WS. f. klass. Phil. 1884. Main- 
tains that C. has used short remarks from his sources 
to expand them into longer speeches. M. Gliick, 
De Tyro ab Alexandro Magno oppugnata ct capta. 
Diss. Konigsberg 1886. A comparison of C. with 
Arrian, Diodorus and Justinus. J. Lezius, De 
Alexandri Magni expeditione Indica quaestiones. 
Diss. Dorpat 1887. The history depends on Arrian’s 
Anabasis, and the fragments of Diodorus, Curtius, 
and Justinus, which last three go back to an author 
of Alexander’s time, perhaps Clitarchus. J. Kaerst, 
Forschungen zur Geschichte Alexanders des Grossen. 
Stuttgart 1887. The best work on the subject that 
has yet appeared. C. Hosius, Lucan wnd_ seine 
Quellen. Kh. Mus. 1893. H. maintains that as 
Seneca has references to Curtius so his nephew 
Lucan may have some. J. K. Fleischmann, ¥. 
Curtius Rufus als Schullektire. Pr. Bamberg, 1891. 
Shows C.’s merits as an author for school reading. 
G. Castelli, ? Hid ¢ la patria di Q. C. Rufo. Ascoli 
1888. Well written, and shows a thorough know- 
ledge of the literature. 





| Part of the Summaries for this Number and the Bibliography have unavoidably been 
held over for the March Number. | 


























Boox IV. 

$1, p. 563. dxddAovbov 8 dv ofpac zepi 
Te paptupiov diadaBety. Kai tis 6 TéAELOS, ois 
euTepiAnpOnoetar...7a TapeTomeva Kal ws 
opotus Te piiocopytéoy SovAw Te kal ehevOépw 
Kav aVIp 7) Yuvy TO yevos TUyXaVy, Ta TE EEN... 
mpocarorAnpicarres...trapabycopeba. Read 
etn for ota, or insert it before it, as we 
have in § 2 évopevov ay «in diadaPeiv, and in 
Str. ii. 1 é&js 8 av ein dadaBeiv. Omit re 
after dpoiws, and put a fuil stop after 


en I AE I TT TI 


avn. 
§ 2, p. 564. ra wepi apyav drowdroynbevta 
A a , Lid 
tots Te EAAnox Tots te GAAOs BapBapors 60 ov 
e A es e 606 > , F 4 
HKov €is nas at doar eEiatopytéov. For daov 
read ocwr. 
$3. 70 pev yeypawerar jv eds ye COeAy.. .vevi 
be él 7 mpoxetmevov peziréov. Perhaps 
aitixa has been lost after yeypawera. 
—— «$4, p. 565. gore dé jpiv ra Dropyy para... 
, / 
bua Tos dvedqv azreipus evtuyxdvovras TOLK LAG 
< > , * Ud / I t 
OS a0TO Tov Toivon.a Pyotr duecTtpwpeva. Inser 
kat before dmefpws and change zoixida into 
puis. 
Ib. ot 8€ Tot xpvcod GvTws yevous TO ovy- 
ves peradAevovtes eipynoovar TO Todd ev 


mention of the x.y.) read dvres, comparing 
$16 dycopev rod ypvood yéevous civar. [H.J. 
calls my attention to 77 Geta dvrws iepopavria 
in § 3, and rijs Bacruxis dvtws 6d00 in § 5, 


tween his own application of the Platonic 
nyth and the myth itself.] We might be 
osed to omit eva, but see i. § 182, 
. 427 Trdtwv eva tov vopobérny pyciv, 
‘NO. LXXVI. VOL. IX, 


The Classical Review 


MARCH 1895, 


CRITICAL NOTES ON CLEM. AL. STROM. IV. 


év 6€ trois Nopos &va tov cuvnvovta Tay 
povotkov.! 

Ib. det b€ Kat 7) ws rovrous (the Stromateis) 
TpooekToveivy Kat tporeperpicxew Erepa, ere 
kal Tots ddov dmotow iw otk loarw dpxa 
THY pepovoar broonpnvar povov. § 5, Badvrréov 
d€ TO peTa Taita HON Kai Tiv Aouriy davrois 
e€evpyntéov. Should we not read teas 
here? C. is urging his readers to do their 
part, not simply receiving directions from 
others, but following up the true path them- 
selves. Put a comma after povoy and write 
e€eupereor. 

§ 6, p. 565. eixdrws obv todd rd yormmow dv 
Oyo TT Ep Mato eurepexopiver THdE TH 
mpaypateia Soynatwv. Should we not read 
o7éppate tov for omepparwy, ‘Great is the 
productiveness which the doctrines con- 
tained in this treatise inclose in a small 
seed ’ 4 

§ 8, p. 566. 4 pot doxed ro aadBBarov & 
aroyns Kaxov ¢yxpareaav alvicoerbar wai rt 
tot éotiv w diadepe Onpiwy dvOpwros. Tovrow 
TE av of Tov Geod dyyeAn codwrepom. For 
ri read 6 rt (from doris), and for re read 8¢. 

§ 9, p. 567. 7 por doxed Kai Mv@aydpas coddr 
pev elvac tov Gebv A€Eyw povov, dra wai ry 
dméatodos év TH pds ‘Pwpalovs trurtoAy 
ypade [els iraxoyy miatews eis wdvra ra MOvy 
yroptabevros] pov code Gog dua 'Inood X puro’, 
éavrov d8...@tAdcogov. Read Adyew for Aéyow. 
The words in square brackets have no 


1 I am obliged to Dr. Jackson for the reference to 
Legg. ii. 658 EB, éxelyny elva: Motear xaddloray tras 
rovs Bedtiorous wal ixavads weradeuptrows répres, 
udrrora Be Hris tva Toy dper® re wal waidels Ssa- 


épovTa. 
pep " 


98 THE CLASSICAL REVIEW. 


meaning or construction as they stand, and 
should probably be placed after Xpiorod. 
Faith and philosophy would thus be coupled 
as human, in contrast to wisdom which is 
divine. Perhaps the disarrangement may 
have been due to a reader who observed that 
the order was different in the Ep. to the 
Romans. The sentence will be made clearer 
if we put brackets before érei and after 
Xpicrod (or rather yvwpicbévros, according 
to the proposed order). [I.B. would insert 
aivitrer Oat after Ivfaydpas. | 

Ib. géouev & otpar Kevravpw [@erradixo 
mAdopati|. The words in square brackets 
seem to be a gloss of the same kind as we 
had in i. p. 342, explaining ras Bacdvov 
NiGov. 

§ 11, p. 568. ovxovy tados rod doBov 
yevvntixos 6 vopos. Read with Sylburg za6ovs 
for the za6os of MS., which Dindorf omits. 
C. does not deny that the Law causes fear 
(z.e. the rational avoidance of evil), but only 
that it produces the passion of fear. [Per- 
haps insert éuzroiv after raGos. I.B. | 

§ 14, p. 569. otros otv poBw 7d apveicbar 
Xpuorov dua THY evToAny exkXriver, va dy PoBw 
pdprus yevqrau. ov pyv ovoe eride dwpeav 
TOYLAG pEVOV Tmumpao KwV Ty miotw? ayary 
d€ mpos TOV Kupiov dopeveotara Todvoe Tod Biov 
amodvOynoerar. Insert od after otv and put a 
colon after yevyra. C. is here contrasting 
the motive of love which actuates the true 
gnostic (tov du’ a&yarns morov as he is called 
in the preceding section) with the lower 
motives of fear and hope. In like manner 
we read (§ 29, p. 576) detv 8 otpau pyre dia 
poor Koldoews pajre dua Twa erayyeAlav 
ddceus, du avo b€ TO ayabov mpoceAnrvbevar TO 
cuTnpiv Oyu. 


© Lb, sp: B70. eit 8 ot év Toke eT’ 


erOupov arobvyncKovcw, ovdev ovTor diadéepov- 
TES €i Kal VOoOW KaTEe“apaivovto. Compare the 
preceding section where death in battle was 
contrasted with death on a sick bed (od 
TpOKOPLOV 7H UXT ovde karopodaKce Gels ola 
rept Tas vomous mdoxovow ot avOpw7or) and 
insert 7 v7] before ei. 

Ib. €i roivuv 7 mpos Gedv Suoroyia paptupta 
éorl, Taga 7 KabapOs roditevoapevn Woy) pet 
eriyvorews TOV Geod 7) Tals évtoAais éraKynKovia 
paptus eott Kal Bio Kat oyw, Orws TOTE TOD 
copatos araXatTyTal, olov aia THV 
miotw ava Tov PBiov amavta, pos d€ Kal TV 
é€odo0v zpocxéovea. Should not we omit 
the article before tais évroAais, putting a 
comma after Wvyy and before pdprus, and 
read dwahAdrrera? I think too that zpos 
should be taken as a preposition and followed 
by a dative 77 é&ddw, ‘ pouring forth his 
faith like the blood of the sacrifice through- 


out his life and especially at his death.’ 
[I.B. would keep to the text, translating, I 
suppose, ‘pouring forth his faith during 
his life, and above all his martyrdom, like 
sacrificial blood.] In the sentence which 
follows, ..onow os av Katadyn 
TATE... EVEKEV TOD evayyeAiou Kal TOD OVOU“aTOS 
pLov, paxdpios ovtoct [ov tiv arr hiv eppaivov 
paptupiav aGAXa THY yoortxiyr] @s KaTa TOV 
Kavova TOU evayyeAtov Tout evo dpuevos Oud 
THS mpos TOV KUpLoV ayaTys, ywoow yap 
onpaiver 7 TOU évopatos ELOnoLs Kal 7 TOU 
evayyeAiov vonois, I think the words in 
square brackets should be placed before 
yvaouw, Which supplies a reason for yyworttKyy, 
while at the same time the nom. zoArevod- 
pevos iS brought into relation with the 
subject ovroat. 

§ 18, p. 572. & yotv tO Gv 7o et Gv 
KatopOovtat kal eis €€w aiduoTNTOS TapameuTeETaL 
6 dia copatos pederynoas evfwiav. Transfer ed 
before the first Gpv, ‘life is perfected by 
living well.’ 

§ 23, p. 574. 
Oatépov 
TOLEL. 


, 
& KUpLos 


6 IlXotros, dyci, tAEOV 
Brérovras mapadaBov tuddordts 
The original preserved in Stobaeus 
has ka@dzep iatpos Kkaxds instead of Garepov 
mieov. Did C. write iatpod tporovt The 
first syllable of tpérov may have been lost 
after the preceding -rpov, and -rov changed 
into zAéov. 

§ 24, p. 574. ev 7G ’AdeEdvdpw 6 Evpuridys 
TeTOLNKEV. €lpyTal ye 7 Tevia Todiay eaye ue. 
To Ovatvxés. Perhaps xai cikdtws may have 
been lost before cipyra. ([I.B. suggests te 
for ye. | 

§ 25-26, p. 575. Kav duly dia duxacocvvyy, 
paptuper Sikalocvvyv TO apiotov Tuyxavel. 
“‘Opoiws 5& Kat 6 kdalwv...d Sixacoodvnv 
paptupee TO PeATiotTw vopw civat KadO. 
Perhaps we should read r@ vopw civar Kado 
Kal Bedtiory. 

Ib. dpotws 8€ Kat tovs kadovs k.7.d. Place 
a comma instead of a full stop before époiws. 
The accusatives depend on the preceding 
pakapicer. 

§ 27, p. 575. After quoting 6 yap ctpov 
(MS. épav) tiv Woyiv arodgce ait C. con- 
tinues 6 Toivey érvywdoKwv...dpapTwddov THY 
Wuxyv arohE€oet aitiv THs dpaptias As 
aréomactat, amoXecas 6é cipyoe Kata THV 
braxonv THV avaljoacay pe tH TioTE 
amofavotcav dé 7H duaptia. C. appears here 
either intentionally or unintentionally, to 
have confused azoXécer with daodvce. In 
p- 679 the form drodvew is used in a similar 
connexion, ypyvat yap amoAvew Tod cwpLaToSs 
Kal TOV TovTOV dpmapTnpatwy TiHVv wWoxyV. 
Perhaps rv should be omitted before dvafy- 
vaca. 


THE CLASSICAL REVIEW. 99 


§ 28, p. 576. etAoyov éaywyyv rd oTrovdauw 
TvyXwpovst Kai oi Piidcodor, ci Tis TOD mpda- 
gew aiTov OUTwWS THPHTELEV abTroYn 
os pykere droheAcipbar aita pond eArida ris 
mpafews. Read with Arcerius oftw orepyrevev 
and for airév, which may have crept iu 
from the preceding airov or the following 
avira, read perhaps dvdyxn. 

§ 34, p. 579. pr pepymvare tH Woyg baw tr 
ddynre, pydt TO oHparti wepiBddr(nre. Read 
with Potter 76 cdpari <ti> repiBddrnre. 

§ 36-37, p. 579. cioi yap rapa Kvpiw Kai 
po Got Kat moval meloves kata dvadoylav Biwv: 
ds yap av défyrat, pool, tpodiyryy cis dvopa 
mpopyrov pucbov rpodytov AjnwWerar...7ddw Te 
a’ tas Kar’ a€lav duaopas Tis aperis, eiyevets 
dpoBas 81a Tov wpov TV oiy bpolwv Tov 

— dpiOpov, Tpds OE Kal Tod ExaoTw Tv épya- 
Tov amodoGévTos icov picbodt, tovtéeate THs 
cwrnpias, [7d éx’ tons Sikarov peprvuKev bid TOV 
kata Tas dkatadAyous wpas épyacapevwv |. § 37 
epydcovrar pev ovv Kata Tas povas Tas dvaddyous 
év katniwbynoav yepav cvvepyovs THs appyTrou 
oikovopias Kai Aetovpyias. Put a comma 
before waAw and a full stop after dperis, and 
insert xara before ras kar’ agiav. Rewards 
are given in accordance with degrees of 
virtue as well as with varieties of profession. 
Transfer the words in square brackets (pre- 
fixing to them the words zpos dé kai from the 
line above) and place them before evyevets. 
The 6a after dyoBas perhaps represents 
diadapBavevrov. The second sentence will 
then (omitting rod before éxaorw) read as 
follows: mpos 8€ Kal 70 éx’ tons Sikatov pep)- 
vuKey Ota TOV KaTa Tas aKaTadAyAovs Mpas 
epyacapevwv, evyeveits apofas <dvadapBa- 
vovTwV> TOY wpOv TOV Ox bpotwy TOV apLOjov, 
ExdoTw Tov épyatav amobobevtos icov pubod. 
In the first sentence of § 37 omit the xai 
before Ae:rovpyias and insert it before xara, 

_ translating ‘ In the mansions corresponding 

to the prerogatives of which they were 
deemed worthy, they will be engaged in 
public services co-operant with the Divine 
(ineffable) Economy.’ 

§ 38, p. 580. eviore yap Bovddpeba...€cov 
Totporat...Kat ovx olot Té €opev row dui Teviav 
a , > fol a s > ’ ba 
7) vooov...e€urnpernoat TH Tpoapere ed wv 
bppwpela pip Svvnbevtes eri téAos dyayeiv O 
BeBovdrjpeba. Put a comma after rpoaperet, 
and for éd’ read d¢’, ‘not being able /rom 
our existing resources to carry out what 
we have desired.’ 

Ib. p. 581. ris abris tysis pebéLovor rots 

buvnbeiow of BeBovdrypévor, dv 4 mpoaipects (on 

iv mAcovertoow ETEpOL TH TEpwovoie. 

Should we not read drepou? ' 

§ 39, p. 581. After quoting the beatitude 

Blessed are the pure in heart, for they 


é 
: 































shall see God,’ C. 


Kara 


continues Kablapor ; otv 


Tus FOPUATUKAS érBupias Kat TOUS 
ayo US dtadoyurpovs Tous cis éxiyrwow roi 
Geov ddixvovjpévous clvac BovAeras, dras proey 
€xy vidov éxurpoaboiw Ty Ovvdpe davtrod rod 


iyepovixev. Perhaps we should read dAAovs 
for aylous. 

§ 40), p- 581. paKxaprot ToWwuy ok cipnvor ow. 
Tov dvTioTparynyouvtTa vOnov TH hpovrjpare ror 
vou ipov...r.Bacceiacavtes Kai é&nucpexrarres, 
ot per’ érotHpns...xataBuscavres els vioberiay 
aroxatagtabycovra. Put a comma after 
eipnvorovoi, and read ot for ol. 

Ib. cin 8 dv } redcla epnvorotnews §. 
pridocovea To cipnvixov dylav te Kal Kadi Tip 
dwiknrw Aéyovoa. Should we not read 
bporoyovoa for Aێyoura | 

§ 43, p. 582. xd ireparobdvys Tov rAnaios 
tAnciov b& ipav 
broAdBys, Ges yap éyyilov 6 odluv mpds rd 
cwlopevov ehexOn, Gavarov AXdpmevos bea Cony wai 
geavtov pardov 7) éxeivou evexev rabwv. [xal 
Hy TLdia TovTo ddeAGds elpyta] bbe dydryy 


ou > ’ . _ 
t ayaTiy, Tov GwrTnpa 


THY mpos Tov Geov rabayv bur ri diay trabe 
cwrnpiav. Transfer the words in square 


brackets after ¢€Aé€y6n, insert fra before 
€\ouevos, and yap before the second &’ 
ayarny. 


§ 44, p. 583. Ocava...ypdda, ‘iw yap re 
VTL TOS Kakols Etwyia 6 Blos, Tornpevrapévoss 
éreita TeAevTOCW, ci ey Tv Gbavaros % Wry), 
[€ppatov 6 Gavaros). Kat HAdrwy dv Paden, 
‘ ei pev yap qv 6 Gavaros Tov wavtds dradAayi, 
kat ta €fys. The words in square brackets 
should be inserted after dra\Aayy). The 
following xai ra ééjs refer to the remainder 
of the sentence in Phaedo p. 107, rod re ow 
paros dua dan\AdyGar Kai Tis abray Kaxias 
pera THS WuyxNs. 

§ 45, p. 583. ot« éorw obv...vociy dxAny 
oluov eis “Awov dépev, dd0i b¢ wodAai «ai 
dwdyovcat dpapria. mtoAvTAavels TovTous, os 
Zouxe, Tors amicrovs duKxwpwdar “Aprroddrys 
x.7.A. Omit xat before drdyovra, which I 
think agrees with ddoi, remove the full stop 
after duapria and place it after roAurAares 
‘there are many roads leading off to Hades, 
sins which cause men to stray in Various 
directions.’ 

§ 58, p. 590. peor) pev oby race ¥) daxAyeria 
Tov peAernodvrwy tov Camvpor Gdvarov els 
Xpurriv wap’ dAov roy Biov. For (onrupor read 
perhaps (worowy. pe this, I see, I am 
anticipated by I. B. 

§ 61, p. 591. rg@ riwrovre roy cmyors 
rapaTtadivar ri érépay. Should we 
read rapadojvact [1.B. suggests rapabeira: 
or mpotabivat. 

§ nee 92. After speaking of the 


Amazons and other masculine Women, cC 
uel 


100 

proceeds 76) yotv af yuvatKes ovdev 
@\arrov tav dppévwv Kal oikovpotot Kat 
Onpevovar kal Tas Totpvas puddtrovor. He is 
thinking of the passage in the Republic (p. 
451) where Plato argues from the use of 
female watchdogs to that of women, ras 
Ondrelas Tov duAdKwv Kuvav ToTEpa Evppvddrrev 
oidueba Set, Gmrep av of dppeves pvddtTwot, Kal 
EvvOnpedvew kal TaANa KoW TPATTEW, 7) TAS HEV 
oixoupety Evdov as aduvdtouvs Ou Tov Ta 
OKUAGKWY TOKOV...TOVS O€ Tovely Kal TaTaV 
éryuéAerav exew mepi Ta Toiuvia. Read ai xives 
ai pvAaxes tor ai yuvaikes. Dindorf says of 
oixovpodar requiritur aliud verbum, and in 
Plato it is no doubt distinctive of the 
female, as opposed to the male; but taking 
it in the sense of ‘guarding the house,’ it 
applies equally to the males. [I.B. would 
insert kuvov after dppevuv. | 

§ 63, p. 592. Kai rs ov parnv Eipiuridys 
TOUKIAwWS Ypaet ; Read zws without a ques- 
tion, ‘perhaps Euripides has some reason 
for the various views he gives of women.’ 

§ 67, p. 593. cadiys iptv éx TovTwv 7 &k 
miotews évorns Kal Tis 6 TéeLos SédecxTar. For 
cagpys read cadas. 

Ib. p. 594. dvdpt aroOvycKew Kadov i7ep TE 
dperns imép Te eAevepias trép Te Eavtov. Read 
<tav> éavrov, or else omit the last izép 
Te. 
§ 68, p. 594. dros dv pay yrrnbertes 
aTOTTALT WOOL TOV GploTuV Kat dvayKaLoTd- 
tov Bovrevpatov. The MS. drorrécwou is 
thus corrected by Dindorf after Sylburg. 
But drorraiw is not recognized by the lexx., 
and in any case seems less suitable than 
GTOTETWOL. 

Ib. ra pev yop GAwv evexa TpaTTopeva HULy 
éxdorore Tpdgaipev av eis exelvous aroP\érev 
FEPWPEVOL...LETPOV WyOUmEvOL TOvTO TO eV 
éxeivois Kexapiopevov. Omit év, or replace it 
by av after éketvos. 

§ 72, p.596. Those alone confess Christ 
aright whose lives witness to their confes- 
sion, év ols Kal avTos dpodoyee Eve tAn p- 

€vos avTots Kat éxopevos tT ToUTwY. 
Dindorf follows Grabe in reading airots for 
MS. adrovs, and would I suppose agree in his 
interpretation utpote qui habitat in illis et ab 
iis tenetur. L.and 8. recognize éhAapBavopa 
in the middle only, with the sense ‘ to seize 
hold of.’ Dindorf in his Index quotes the 
passage under the heading éveAovpevos, and 
changes the MS. spelling éveAnppevov in 
Paed. ii. 81, p. 219, reading otpdépacw 
éveAnpévov. Should we not in our text take 
it as the middle of éve.Aéw, keeping the ace. 
avrovs and translating ‘ having incorporated 
them (lif wrapt them up), into himself’? 
1t will thus answer to the phrase which 


THE CLASSICAL REVIEW. 


follows immediately, dpvotvrat de avtov ot p17 
ovtes ev adt@ and to Joh. xiv. 20 bets ev enol, 
Kayo ev tuiv, Whereas the tets év éuot has no 
equivalent here if we accept the reading of 
Grabe and Dindorf. 

§$ 73, p. 596. éxetvo O€ ovk éréearnoer ott 
k.T.A. Read éxe(vw ‘he did not attend to 
this point.’ A 

Ib. duabeots bé dportoyoupevy Kat padvota 7 
pnde Oavarw tperomevn tf EVGA TAaVTwY 
tov Talov...amoKxoTiyv movetrar. For éva ravTwv 
read évy dravtwv, comparing Epict. Diss. i. 
22,33 av aitovs of Tpdes py aroxtetvwow ov 
py amobavocw ; Nat, add’ obx td’ ev wavTes. 

§ 75, p. 597. redelws bpoAoynoas Kat Tats 
evtoAats kal TO Ge@ 1a. TOD Kuplov, Ov dyaTyTas 
ddeAdov éyvipirev oAov EavTov eridots O14 
tov Oeov. For dua tov Oedv read dv’ airod 
TO Geo, 

§$ 76, p. 597. ‘In warning us to flee in time 
of persecution, our Lord does not imply 
that death is an evil, but He bids us avoid 
giving offence,’ TpoTov yap Twa TporayyéAXet 
éauvtov mepuiotacba. Hervetus translates 
denuntiat ut sibimet caveamus, but can 
wepuictacGar bear this force? In later Greek 
it means ‘ to avoid,’ ‘to keep at a distance,’ 
implying e.g. such an object as Oavarov. If 
the letters were getting faint, Oavarov might 
be easily changed into éavrov. [I.B. would 
read airév. | 

§$ 77, p. 597. otros 8 av ein 6 py TE pe 
oteA\AOpevos Tov dSwypyov. Hervetus 
translates qui non vitat, but there is no 
evidence of such a use. In the passage 
quoted by Potter (Strom. p. 871) ot repurted- 


Aopevor Simply means those who take care. 


of themselves. Perhaps we should read 
mepuctapevos. [1 am inclined now to prefer 
brocreAAOpevos Suggested by I.B.] 

$ 78, p. 598. ef Kyderar tyav 6 Geds, TH 
dnote OuoKeaGe Kal poveverbe, 7) avTOs tpuas eis 
Tovto exdidwow ; Puta mark of interrogation 
after doveverGe, as in Potter’s edition, 

§ 79, p. 598. Kav py ddckOpev, GAN ws 


ddikovow Huiv 6 dukaoTys &Popa. Potter 
suggests éfopya, perhaps édedpever. [Here 


too I think I.B.’s suggestion, idopyet, sup- 
plies the true reading. | 

Ib. atrovs te odtw modutevopevovs Kal TOUS 
dAXous Tov opoLov aipetabar Biov 7 poaTpe 
mopmévovs. Read rporperopevors. 

§ 80, p. 599. ré yap Kai adicotpeba ws 
TpOs Mas aVTOVS Pavatw arodvopevor TpOs TOV 
xvptov ; For ws I should prefer ro ‘so far as 
we ourselves are concerned.’ 

Ib. ci 6€ & HPpovotpev yxapw cicdpcba 
Tois THY ahoppyvy THS TaxElas arodnpias 
Taper xnevols, ei Ov aydarnv paptupotper, et bé 
py, patrol tiwes avdpes etvat tots moAdois 


ae 


ee eee ee 


a 




































pera) ey Pips aA eweser 


THE CLASSICAL REVIEW, ol 


edoxodpev jets. €t ydecav d& Kai abrol riy 
aAnGeav, wavres nev dv ereridwv TH 650, éxAoyi) 
‘ > x” > . . . 

8¢ ovx dv Hv. Potter is right, I think, in 
reading ¢dpovodpey (which was no doubt 
altered to suit paprupotuev), omitting the 
comma before ¢atAo, and changing the full 
stop after jets intoa comma. Put a full 
stop also after paptrupotwev and omit ei before 
joecar, translating ‘If we had not seemed a 
bad sort of people to the majority, and they 
had themselves known the truth, all men 
would have pressed into the Way.’ 

§ 81, p. 599. dypi yap +6 broco tbroriz- 
Tovar Tats Acyopnevats OAiverw Hror jwapTnKoTEs 
év GAXots AavOavovtes Traicpacw eis TovTO 
ayovtat TO ayabov xpynororyt. Tov TEpidyovTos, 
dda €€ ddXwv OvtTws eyxadovpevo, iva 41) 
Os KaTaduKo eri KaKots Gporoyoupevors TAIwWCL, 
pnd Aowdopovpevor ds 6 porxos 7) 6 hoveds, GAN’ 
67 Xpiotiavoi repuxdtes. For 76 read Gru. 
The yro before ypapryxdres suggests another 
alternative ; and if we compare § 83 rpoapap- 
tTHoacdv yor Tv Wyn ev Erepw Biw THv 
KoAacw bropevery evtadda, Thy pev éxexTijy 
éxitipws Sua paptupiov, and § 88 KoddleoGar 
pev tov pdptvpa ba Tas mpo Thode Tis 
evowparacews apaprias, I think it will be seen 
that some such phrase as the following has 
been lost, <év dAAw Biv i) evratOa> ev adrAors. 
Perhaps 6vrws may represent an obliterated 

evdopaptupovvtwy or diaBaddovrwv. I am 
disposed to think that the last clause read 
originally as dre porxds 7) dpoveis, GAN ds 
Xpurriavot. If dru had got corrupted to 6, a 
marginal correction might be thought to 
refer to the later és. Just below put a full 
stop after doxeiv. 

Ib. p. 600. weloerar ds Exaa XE kai 70 
vyruov. Should we not read zdcxer? I see 
no reason for the imperfect. 

§ 82, p. 600. as ody To vyTLOV ov zpon- 
papryKds 7) évepyas pev odx HpapTnKos ovder, év 
éavrG 8€ TO dpaptiaar exov, éxiv iro 
Brn6j 7d wabeiv evepyerciral te moda 
kepdatvoy dvckoAa. For apaprjoa read dpap- 
Tytikov, as we have just below ‘Exwv pe ev 
aut ro dyaprytixov. Omit re which is a 
mere dittography of -rat. 

Tb. rabrd eraber gudepas 7G vytw. Read 
Taira. 

§ 84, p. 601. ép& rotvw atrov éxi toi 
kparnOévros SporoyyTov, TOTEpoy apTUpHTEt... 
4 ov. For épa read épwrd, the last syllable 
having been lost before the following roc. 

Ib. i 8 kat tis droBdoews Kal Tod pu) dev 
KodacOjvat totrov pyoe THv drdeav TOY 
dpvnoopévev éx mpovolas aKwv T poo LapTUpHT et. 


C. is here arguing against the view of 


Basilides, ‘that martyrdom is the provi- 
dentially ordered punishment for ante-natal 


sins, Martyrdom, he says, may be avoided 
by the denial of Christ : if such aA result 
is providentially ordered, then Providence is 
responsible for the perdition of the rene 
gades, The text is evidently incomplete 

insert rode before THs, and rp wpcvesas aitiay 
before TH drwAaay. 

S 85, p. 601. we tp a Cwy yap 6 ba Boos 
eidws pev 5 dope otk elds BR a irropevrotper, 
adXa droodoat...BovAdpevos «al irayerbas 
€QuTW TWetipaler 


lransfer rapd{aw and 
TrEeipacet. 


The former was probably changed 
in consequence of the preceding Gedfew. A 
little below put a full stop before roo dn ") 
TLOTLS ; 

S$ 86, P- 601. ef be... pépos tx rot Acyo 
sega GeXyparos Tov Geod tradrdapey To Wye 
THKEVaL aTavTa...€TEpov € TO pwndevds éxchupeir, 
kat tpitov puceiv pnde ev, | GeArpare rod Geod) 
kai koAacves Ecovrar It is difficult to 
see how the conclusion (that punishment is 
ordained by the will of God) can be derived 
from the preceding analysis of God's will 
into universal benevolence, absence of desire 
and absence of hate; indeed just below it 
is asserted that such a conclusion is impious. 
Perhaps we should put a mark of interro- 
gation after écovra, and transfer the words 
in square brackets to the end, inserting 
7 after pndé &. 

§ 87, p. 602. (We must not suppose that 
God is himself the efficient cause of perse- 
cutions) dAAG pi) KwAvew Tods dvepyoirras... 
kataxpnobai te cis xaddv Trois Taw dvarTiow 
ToApnpacw. For re read dé. Just below in 
maidevtixys TEXVNS THS TOLGDE ALTHS ofoys 
mpovotas, read roatrys for rouide airis. The 
corruption originated no doubt in a super 
scribed correction -aade. 

§ 88, p. 602. 1) mpdvor be ci wai dwd rot 
dpxovtos, as pavat, xwciobac dpyeras. Here 
dpywv is a technical term of the Gnostic 
theology (cf. Str. ii. p. 448), and we should 
probably read ¢aciv. [I.B. suggests darepes 
for ds pdvat. | ; , 

§ 89, p. 603, epi piv Tovrea rods 6 Asyos 
Soov ev toréipw cxordy droxcoeras. For 
Scov read $y, the first syllable is merely a 
dittography of the preceding. 

Ib. ro Bi ddopor yeros. 
as in § 91. Lad 

§ 90, p. 603, drei d8 rd dawopevor abroy 
otk tor  éx perdryros Wuyi fpxerar FO bem 
bépov, kat rour fer To épov , ror 
duadtpovros mvEvparos [xai «aff oAow) © ‘~ 
rverrat TH Woy TH cixove rot svevparos. For 
atrod read atro, transfer ro bsadipor before 
rov biadépovros, omit the following «ai and 
the words in square brackets, which have 
slipped in from the xai xaééAov of the next 


Kead dsagddpor 


102 


line. Translate ‘since the merely apparent 
(i.e. the Demiurge in his first shadowy form, 
as described before, when he is as yet 
merely an unsubstantial image of the true 
God) does not really exist, the mediating 
soul comes, @.e. the excellent inbreathing of 
the excellent Spirit, which is breathed into 
the soul, the image of the Spirit.’ 

§$ 91, p. 604. ei éri TO Katadioo Oavarov 
aduxvetran TO Sia épov YEVOS; ovx 6 Xpioros TOV 
Oavarov Katnpynoev, ei pay Kal adds avTois 
bpoovoros Nex Gein, ei 6 eis TOUTO Kar Tp 
yyoev os ye TOU Stapéepovros antec Oat evous, 
ovx ovTot Tov Oavatov KaTapyovrw. For 76 read 
7@, for eis Todro read cis rovrov, translating 
‘if the Elect come for the purpose of de- 
stroying death, Christ did not abolish death, 
unless he should be said himself to share 
the nature of the Elect ; but if he by him- 
self abolished death, without having (so as 
not to have) any part in the Elect, then 
death is not abolished by these. Just 
below for xara tHv Tod Sdyparos aipecw we 
should rather have expected kara 7d ddypa 
THs aipeoews, but it may be enough to insert 
tovtov before rod ‘according to those who 
hold this dogma.’ 

Ib. cin & Gv Kat 6 Kvpios apeivwv Tod 
Synprovpyod Geov, ov yap av mote 6 vids TO 
marpl diadiAoverkoin kal Tatvta év Oeots. For 
ely ® av read ei de, and insert adAa before ov 
yap, decline ‘even if the Lord is superior 
to the Creator, still there could be no 
rivalry between father and son, and that in 
the case of Divine Persons.’ 

$93, p. 605. ro duaprdvew evepyeta Ketrar 
oixk ovoia. Insert év before evepyeta and 
ovoia. The former would easily be lost and 
the latter would naturally follow it. [I.B. 
suggests évépyeia and ovcia. | 

[$ 94, p. 605. Kara qvouxyy roivuv dpeéw 
XpHoTEoV TOIs<py>KexwArvpevors KaAds. I.B.] 

§ 95, p. 606. rods d6uoroyodvras wey EavTods 
elvat TOS TOU Xpiotod év d€ Tots TOD dia- 
BoXrov kataywopevouvs épyos. Dindorf after 
Sylburg inserts év before the former rots, 
but this does not correspond with the fol- 
lowing év épyos. If we omit rots as having 
crept in from the following clause, we get a 
more satisfactory sense, ‘those that profess 
themselves to be Christ’s.’ 

§ 96, p. 606. ra éveorara év ois eopev Karta 
tov Tov Piov xpovov &s Tod pev oTpaTUdToV % 
é€Amis Tod éurdpov Sb 7d Képdos. Perhaps 
Nefa has been lost before édzis. 

§ 97, p. 606. After quoting ovK ev TaTW 77 7 
yaors C. continues eici d€ of pact TV Tept 
tov eidwrofitwv yvdow oik év maar dh Ep Etrv. 
A more suitable word would be éxdépew. 


[cvuppeper. I.B.] 


THE CLASSICAL REVIEW. 


Ib. kav pdockwor ‘wav 7d ev paxéeAdw Twodov- 
pevov ayopace Oet,’ Kata retow emdyovtes TO 
pndev avaxpivovres, em’ tons TO <py> avaxpi- 
vovtes, yeAdoiav e&yynow tapabynoovra. Din- 
dorf after Potter inserts px before the 
second dvaxpivovres. A comparison with 
Strom. 1. p. 370 will show that the MS. is 
right. There we have ovxt, eudpavey o Geos 
THY copiay TOU KOO LOU ; er’ ions TO * pwpav 
édevée.” So here, if pydev avaxpivovres forms 
part of an interrogative sentence ‘are we 
to buy without asking questions?’ it is 
equivalent to an imperative ‘ask questions 
when you buy.’ 

Ib. (Buy freely from the shambles with 
the exception of) trav dyAoupévwv Kata Tv 
emioToAny THY KaGoALKHY TOV aTooTOAWY amdav- 
Twv ov TH EvdoKia TOD aylov mvEevpaTos, TH 
yeypappmevy pev ev tals mpdkeot THY aro- 
otodwv, Stako pia belo 7 0€ cis TOvs TITTOUS 
8’ avrod: diaxovodvtros TlavAov. For the 
datives read accusatives agreeing with 
émioToAnv, comparing my note on Sir. iii. 13 
for a possible instance of a similar corrup- 
tion. 

Ib. p. 607. wavra oréyomev iva py eyKomny 
ddpev TO ebayyeAiw Tod Xpicrod: nro dopria 
mepidyovres S€ov eiA’tous eis madvTa civat, 1 
brddetypa Tots Pédovow eykpateverGar ywvo- 
pevol, 7) OLKOOOMOVPMEVOL Eis TO aNdaS 
Ta mapatieueva, eoblev Kat ws ervxev Spirety 
TH yvvakt, padvora 6éx.7.A. This is a com- 
ment on St. Paul’s words ‘have we not 
authority to eat and to drink and to lead 
about a wife?’ OC. instances two ways of 
causing a stumbling-block, either by carry- 
ing about impedimenta (whereas the 
Christian soldier should be expeditus), or 
by countenancing those who were un- 
willing to deny themselves, and who would 
be thus trained (lit. ‘being thus trained ’) 
to eat greedily what is served up and to put 
no check upon themselves in their inter- 
course with their wives. Put a comma 
after Xpicrod, and a full stop after yuvacki. 
Read oixodopovpevors to agree with ots 
OéXovew, remove the preceding py and place 
it before @Aovow. Dindorf (following 
Lowth) places it before dndés, but this is 
from failing to observe that oixodopéw here 
has a bad sense, as just below, iva py KaKxs 
oikodopyfy, and in 1 Cor. vili. 10 4 ovved- 
dynos avtod aobevovs dvTos oikodopynOynoerat eis 
TO TA ciOwAdOuTa eoGiew. 

§ 98, p. 607. wavra ovv doa rovetre cis OdEav 
Geod woueire? 00a UO TOV KavOva THS TiaTEWS 
moveiy émitérparrat. Sylburg (in his Index 
s.v. t7d) makes jrd dependent on zroreiy, 
translating regulae subjicere, but surely 
voy must be interpreted by the pre- 





































THE CLASSICAL 


ceding zrovetre. 
Kavovos. 

§ 103, p. 609. After quoting from the 
list of O.T. worthies in Heb. xi. CG. con- 
tinues dre pév oty pilav owrypiav éye ev 
Xpiotd tov dtxalwv cal jpov cadds pev 
eipnxev tpotepov. For dixaiwy read dpyaiwv. 

§ 110, p. 613. rAnOous eX€ov repréexec bar 
Tov éAmilovra...Aéyer. Insert izd before 
mAnGovs. Dindorf follows Sylburg in read- 
ing 7A76os. 

§ 112, p. 614. eirep otv Kai picbd zpoc- 
Soxwpevw eukpoToOv Ta xeElAn eis papTupiav 
Kuptouv bporoynow KUpLoV, KOLVOS eit aVvOpwrros, 
ov ywooxwv. Insert éri before picbd. The 
strange phrase emxpotév yx«/An is perhaps 
an allusion to xaAxos 7xov of 1 Cor. xiii. 

§ 114, p. 615. After quoting Matt. v. 28 
C. continues ov WAnv thy erOvpiav jElov 
kpiveoOa, dAda eav TH erOvpia TO Kar’ ari 
epyov...ev €avty exreAjrar 7 yap ovap TH 
davracia cvyxataxpytar 7 5 n Kal TO copatt. 
For 7 read 7 and transfer 737 and 7, (‘ where 
passion rises to such a height), it has already 
been affected by the imagination in dreams 
as though by the actual bodily presence.’ 
Doubtless the corruption arose from a mar- 
ginal correction, 75) being referred to the 
wrong y. [A simpler and better emend- 
ation is that of 1.B. reading 7% for 7, as 


Perhaps we should read rot 


though it were 7 6vap 7TH $. TVyKaTa- 


pOMevy cvyKkataxpyrat. ; 

§ 115, p. 615. rporAaBovons svap, Tis ex 
Ovpias [ryv maida] map’ éArida Kopecbeis 
HKOVoaY THY Epwpevyv KaTa TO TEeTayLEVOV Eipyel 
Tis €toodov. The words in square brackets 
were probably lost from before rijv épwpevny, 
and being afterwards added in the margin, 
were inserted in their present position by 
one who was unfamiliar with the intransitive 
use of zpoAaBovons. 

§ 120, p. 618. dact dé kai tas "ApyoA.kas, 
Hyovperns abrav TedeoiAAns...Xraptuitas Tous 
dAxipous Ta rodéuia. havetoas povey tpevarbat 
Kal éxeivats To does TOD Gavarov TepiTouy)- 
cacbo.. For éxeivais read éavrais. Potter 
would insert airjv, but would not this 
require the active repitoujoae ! 

§ 124, p. 620. ra pev otv adda cipyew 
Svvarai tis mpoorodcuav, To 8 ed’ Hiv 
ovdapas, ob’ &v pddiora evioraro. For av 
read ci. Possibly we should read dvvair’ div 
for dvvarar before. 

§ 137, p. 626. kdv rus dyabouvpyotvte air@ 
évavriov te dravryon, as ayadiy THY avTe 
pucbiav aurvnoixdkws mpojoera. For dyabiy, 
which has perhaps slipped in from the pre- 
ceding éyuboupyotvr, read dmyyopeupevyy. 
Dindorf follows Lowth in prefixing ov to 


 dyabijv. 


REVIEW. 103 


§ 138, p. 627. After describing the pertect 
Christian as ovK. éyxparns ett dAX’ éy abe 
‘participant of the divine nature, 
C. goes on to say é7ray b¢ cy fe TOUry Tu 
EVEPYETLKOV, diow dyabot pipenoreras : 
[dpOévras perarcOjvar, dAAd 
KéeoGar of dei. 


= : 
atraveias, 


ov Oct 64 
Badiovras ddu 
| rovro yap cart TO éAxvobrva 
bro TOU Tarpos | but raons THs orevns buAGvras 
6d00, TO aévov yeverba ri divapew rips Xaperos 
mapa tov Oeot AaBeiv dxwAvtws dvadpapeiv. 
The ideas do not hang well together. Potter 
noticed that dd TAOS THS oTevns...d600 Was 
unsuited to éAxvoG@jvar tro rod watpés and 
must be joined to the previous clause Sadi 
fovras «.7.A. On the-other hand the last 
clause, as well as éAxvo@jvat, would agree 
well with the expression dp@évras perarefyvas, 
the whole being descriptive of the new 
nature of the perfect Christian as opposed 
to the lower state of the éyxparys described 
in the phrases already noticed. I think 
therefore we should read ot dei 6¢ BadiLowras 
aduxéoOar of Set du maons THs orevas ber 
Oovras 6800, dAXr’ apbevras petarefjvar’ TovTo 
yap eotw TO eAxvoOnvar bro Tov TaTpos, TO 
aéwv...dvadpapev, translating ‘the journey 
to our goal is not all on foot through the 
narrow way; we must be lifted up and 
translated. For this is the being drawn by 
the Father, viz. the being made worthy to 
receive from God the power of his grace to 
run upwards without hindrance.’ Compare 
Potter’s note on dwafea Paed. i. p. 99. 

§ 142, p. 628. Kat dy Kai u] eixav tow Bar- 
risparos «tn dv Kal &k Mwiotws rapabedo- 
peéevn Tots Tomrats bd¢ ws. Omit # (often 
confused with xai, see my n. on i. 7, p. 319) 
before eixéy and before éx. 

§ 145, p. 629. Kai 5 ye 'Exixoupos dducciy 
emi xépder tut Bovrerbal dnote toy Kar’ abrov 
copov: miatw yap AaBetv repi tov Aabeiy ob 
Svvacba. Insert od before dyai, otherwise 
the succeeding clause would have been con- 
nected by dAAa instead of ydp. [Here I am 
anticipated by 1.B.] . 

$ 145, p. 630. amis 6€ Gporvpes Kas 
4 THS eAridos drddoais TE Kal droxaraoraats. 
Read dpdvupos: * Hope is here equivalent 
to the restoration and regeneration hoped 
for.’ + a er : 

§ 147, p. 631. rots be wba ciypyeroas % 
Suoypyotas TpoTayomEevors Tey fe 
dméxovrat, Tov 3 ov. Perhaps rporpepoperos 
may have been lost after et oe 
[I.B. would read simply zpocayoperot 

[ Jo. pucrarropLevot, ravru. Put the comma 
after ravra. I.B. ; 

Ib. xiv 1G Soxeiy murras dvactpeperras +4 
xpiow €xovrw dvéowov. This clause shou 
be preceded and followed by a full stop. 


104 


§ 150, p. 632. 6 pev oty avOpwros ams 
ovTOS Kat ideay TAdToeTAL...6 b€ Tus avOpw- 
mos kata TUTwow. For obros read ovtws. 

§ 151, p. 632. od Tavry dpoBos (6 Ge0s) 7) i 
Ta dewa exkALve...ovTE yep av TEpiTET OL TW 
dewd 7 Tod Geod picts, ovre Pe byes 6 GOeds 
Settdav. Insert with Potter ov before 
éxxAiver, and change devyes into devyor, and 
detAiav into deAta. 

§ 153, p. 633. 


S5% X\ S02 See : > \ 
QUTLKa TO Eh Ylv EoTLV 
> 
ovmep ér 


ions aitod Te KUplol éopev Kal Tod 
GVTLKEMEVOY AUTO, OS TO PiAocodely 7) Mi), Kat 
TO morevew 7% amorteiv. Read rod for 70 
before dirocodety and morever. 

Ib. of 7 ad Kodralopevor evexev TOV yEVvo- 
pévwv avTots dpaptnparov é€m’ avTots povots 
KohdLovrat, api Abe yep TO. yevopeva.. -apievrat 
yoor Tpos TOU Kuplov ai ™po THs TlorEws, ovx 
iva py Oo yevomevat GAN’ ws py yevopmevat. 
Perhaps we should insert dapria. after 
miotews, unless there is some corruption in 
the dpaptnuarwv and airots povos of the 
preceding sentence. 

§ 154, p. 634. mpdrov pev tv’ attos apeivov 
avTov yevntar 6 madevdmevos, cio éTeELTa 
orws «.t.A. Omit eis before ézera, as a 
dittography of the preceding -os. For a 
similar corruption cf. my note on iii. p. 514. 
[1.B. supposes something to be lost between 
eis and ézeura. |. 

§ 155, p. 634. eikdrws ovv Kat TAdrwv tov 


Tov idedv Gewpyntixov Oeov év avOparos Cyoeo Oat 


pynat, vois dé xdpa idedv, vods dé 6 Oeds. TOV 
aopatov Geodv Pewpytixdy Oedv ev avOpurrots 
favra «ipynxev. Perhaps we should change 
the first d€ into yap and dyou into dias, 
making the following clause (vois...eds) a 
parenthesis, with a comma after Oeds. Tod 
has probably been lost before doparov. 

Ib. orav yap Yet yevérews imeavaBaca. 
Kal’ éauTnv ye a kat opidy Tos eldeow...otov 
adyyehos On yevouevos oly Xpiotd Te eorat 
Oewpytixds dv, «.7.’. For ye read te, which 
has been wrongly inserted after Xpior@. 

§ 157, p. 635. was vids dhAovyevi)s darepi- 
tl Kapoia Kal darepitpntos | €CTL capt, 
TOUTECTL axdlaptos TOMATL Kat TVEVPLATL, ovK 
cioeAevoeTar eis Ta ayia. Omit éori, as in 
the original (Ezek. xliv. 9). It was doubtless 
the insertion of a scribe who did not under- 
stand that the verb civeAcvoerar belonged to 
VLOS. 

§ 158, p. 635. (Of the tribes of Israel those 
were accounted the holiest) ai cis dpyuepets Te 
kat Baoireis Kal mpopytas xplovgar As 
we are not told of any tribe which had the 
right to anoint, except the tribe to which 
the high priest belonged, Lowth suggests 
that we should read ypicbetoa, which would 
then apply to the priestly and the royal 


THE CLASSICAL REVIEW. 


tribes; but there will still remain a diffi- 
culty, as the prophets were not confined to 
any tribe (cf. v. § 41, p. 670). 

§ 159, p. 636. After speaking (in refer- 
ence to Ezek. xliv. 26, 27) of a purification 
(ka0apifeoOar) which lasts seven days, and a 
propitiatory sacrifice (id ao pov) on the eighth 
day, C. continues réAevos 8 otpar oddauriite 
4 Ova vdmov Kal mpodytav eis TO evayyédvov 
miotis ihLews Kal H OU braKons TAacns ayvela 
oiv kal TH arobéca TaV KoopLKOV cis THY €K 
THs dTWoNaAVTEwWS THS WuxNs EdXApLOTOV TOU 
oxyvous drodocw. Put a colon after riots 
and read idacpos 6€ for (Aews kal, and drro- 
Avoews for arotavcews, comparing Arist. 
Resp. 17, 8 avaicOynros } THs Wuxns aarddvats 
(év yypa). In the perplexing sentence which 
follows, it seems to me that the words ér 
yop tporys ethaBe Kai THs EBdduns arrerar 
aepidopas Should be removed from the end 
of the § and placed after «ire kat 7 dmAavys 
X“pa 7) mAnoidLovea TO VONTO KOTpw dydoas 
Aéyorto: the eighth sphere (that of the fixed 
stars) is in close proximity to the furthest 
planetary sphere, and therefore not entirely 
removed from fear of change. 

§ 160, p. 636. The just man will depart 
from this world yvpros...duaptias Kat Tod 


e / a 3Q/ , 3 a 
€7 O[LEVOU TOLS adikws Bidcaci Qet 8 OUS 
eidwAov. Theallusion, as Potter has pointed 


out, is to the Phaedo 81, where it is said of 
the sin-polluted soul that it still retains 
something of an earthy and corporeal nature 
on its separation from the body, and hence 
continues as a visible ghost to haunt the 
tomb where the body lies: it is only the 
pure soul which is dedys. Read therefore 
yewoods for dedots and put a full stop after 
eidwdAov. In the next sentence rotro yap jy 
TO eipynpevov ‘ eay py OTpadevtes yevryobe ws TA 
waidia, KaGapol pev...ayior O&...0€LKVUVTES 
OTL TOLOVTOUS Has evar Bovre€TaL, for deckvivTes, 
which was naturally assimilated to xafapoi, 
read dexviv agreeing with 70 eipypevor. 

§ 161, p. 637. pi wrcovenreiv &v Oat épy. 
Omit éy, as caused by dittography, and 
read Barépou ‘not to take advantage of his 
peige bea 

§ 162, p. 638. 6 Beds dé a dvapxos Spx Tov 
dhov TavTe ne) s apxis TOUNTLKOS. a wey ouv 

éortiv ovo. apx7) TOU TOLNTLK ov TOTOU, 

KaOdcov éotiv Tayabdv Tod HOiKOd, 9 8 ad éori 
vous Tov AoyiKOd Kal KpitiKoD Torov. Put a 
comma after dAwv and read zavreAds. For 
moiytixod, Which has slipped in from the 
preceding zroyrixds, read dvorxod (with I.B. 
in J. of Phil.) and add 62 after xabocov. 

§ 165, p. 639. ai ayaa rpages as apelvous 
TO KpelTTOVL TO TV EVMATL KVP LH TpocaT- 
Tovrat, ai d€ PiAydovor...7G HrTove TO Gpapty- 





eth pores rier AS 


THE CLASSICAL REVIEW. 


tuk@. For wvevpati kupiy read rvevpatixd, 
the last syllable having been wrongly 
expanded by the scribe ; compare my note 
on ii. 89. 

§ 166, p. 639. 
puKOV 
KOO PLOV. 

§ 169, p. 641. riv otveow dkonv eiriv Kal 
otpavov Ti TOV yvwaTiKod Wuxi, THY otpavod 
kal Tov Oetwv Oéav éravynpyyevov, Kai tavry 
‘TopanXitnv yeyovéevav' Eurradw yap air odv édo- 
pevov tiv apabiar...ynv eipykev. Insert EXomévov 
before yeyovéva: and read ad tov for airov. 


> , 

eTieAOVpLEvOS Kal KO C- 
x , ” , > 
tov tomov eva xatadtve. Read 


SOME EMENDATIONS OF 


Soph. fr. 179 Nauck?. 
yovaixa 8 é&eovres 7) Opaooe yevev 
Te ws TOV pev Ewrov ypadiots evnppevors. 

Perhaps écrotv 6 ewrov ypadidios éverp- 
pevots. A. description of a woman picking 
her teeth with a stylus: quae turbat 
masxillam dentemque hesterno cibo fartum 
stylis insertis. 

Soph. fr. 215. 

‘Is not trogpos a mistake for imddopos ! 
The substantive irodopa was used in the 
sense of a hollow passage or pipe, and then, 
medically, of a fistula: trddopos as an 
adjective had a similar meaning. This 
would explain Erotianus’ vrodpov xpvdaior, 
and the scansion would be possible in the 
corrupt verse cited by him from Sophocles’ 
Erigone viv 8 «ip vmodpos e€& aitav ews 
*Amodecev te Kaitos é€arddero. It would 
also suit the passage cited by Erotianus from 


Hippocrates wrodpov kai éyov epi airo 
Gadapas. 


Eurip. fr. 303. 

6 yap ovdervos exdis 
xpovos dixaious érdywv Kavovas 
deixvucw avOpuirwv KakdTyTas 01. 

Dr. Blaydes (Adversaria in Tragic. G'raec. 
Fragm. p. 318) writes ‘hoc non intelligo.’ 
I think the meaning is: Time existing 
from the immemorial past brings up rules 
of right and wrong which act as tests by 
which I am able to distinguish bad men 
from good. 

Eurip. fr. 401, 3-5. 

Tapa T eArioa Kat Tapa dikav 
Tovs pev dz olkwv 6 évarimrovTas 
5 drap Geod, rovs 0 edtvxotvTas dye. 
mapa Ségav is an old and very probable 
emendation of zapa dicav. For oikwy 8 éva- 
mirrovras I suggest div aimrovras. The 


105 


Swi. p. 641. TL... BovAdpevor pera wis 


y P 9 Tet: 
evxerOar xeXcvovow;...drc biuxalas e Bo. 
?- . > ‘ a © ’ 
Aovro elvat tas ebyas is ok dy tis alderbecy 
toveicbar moAAGv aorvedédtwv. I think we 


should either change éBovAovro to ITyouvTo 
or das into ofas. 

Sok. p. 642, eyo dé ay evaipny TO Tv pa 
Tov Xpirrod wrepOoal pe eis tiv ‘lepovoaAiyp 
mv éunv. For éuyv read xawyy, the first 
syllable of which in its contracted form was 
liable to be changed into e. 


J. B. Mayor. 


THE GREEK TRAGICTI. 


metre of 5 is uncertain, but drep Geor 
‘abandoned by the god’ and doe are 
plausible corrections. 
Neophron /r. 3. 
TéAos yap avdtos €xbioTw popw depers 
Bpoxwrov ayxovnv exurmacas bépy. 
aitov exbictw, aitex$iorw are mentioned 
as variants. In this variation I trace a 
combination of airds aitov, and would write 
the passage thus : 
téXos yap abtos abrov éxioctw popw 
pbepeis Bpoywrov ayxovnv érurracas, 
omitting depy as a gloss. éepeis is 
Elsmley’s conjecture. It is also possible 
that éri should be deleted, and ordaas dépy 
written. 
Achaeus 9. 
(A.) May ’AyxeAdos (Blaydes “AxeAgos) jv 
KexpapLevos TOAUS ; 
(B.) ’AAN ove ActEar rade (Blaydes rade) to 
yever Gps. 
(A.) Kadds pev oty dye oKvby mel. 
Possibly xadds peév otv doreiov ds SxvOy meeiv. 
as dvayjoavra in fr. 20 is written for dvaery- 
cavra. ‘Immo pulcre urbanum erat, ut- 
pote Scythae, sic bibere aquam mixtam 
uino. A Scythian drunkard would mix 
no water, but take his liquor neat. 
Achaeus /7. 19. 
tov Srapriarny yparrov KipBw ev durA@ EvAw. 
Read yparroxupBw év dirA@ | EvAw. 
Chaeremon /r. 10. 
v6’ ai piv airay els dreipova otparov 
dvOéwv doyxov éatparevaay 7/Sovais 
Onpwdpevat...ovTa Acpavev réxva. 
Possibly Bpvovra which is often constructed 
with a dative. 
Chaeremon /r. 12. 
roAAiy érapav Kimpidos eloopav Tapiy 
axpaure Tepxalovoray oivdvOats + xpdvov, 


106 


It is ustial to consider ypdvov corrupt, an'l 
Meineke’s éuod has found many believers. 
Of these I am not one, and even doubt 
whether ypdvov may not be defended. 
Chaeremon is one of the most artificial 
poets of the later kind of tragedy, meant 
like Philip van Artevelde, and Bothwell, not 
for acting, but reading. The poet, speaking 
of maidens in the maturity of their charms, 
describes them as fruit darkening with 
the perfect vine-blooms of time, 7.e. with 
the hue or tinge that maturity brings to 
female beauty. Such language belongs to 
a late stage of poetry, and may be com- 
pared with Propertius’ mortis /acrimis, tears 
of death=our tears when dead, or tears 
after death. 

Chaeremon 13. 

Kopatow opov cwpat’ evavOn poda 

elxov, TUOnvnp’ Eapos extperéoTarov. 
owpar is, I think, wrong, and I doubt 
its beizg a corruption of oréupar. It is 
more likely to stand for ypwpar’, a word and 
idea of which Chaeremon is particularly 
fond. Roses are called the season’s diverse 
hues, because their colours vary with the 
successive times at which they flower. Or 
‘Qpév may be personal, the Hours; roses, 


THE CLASSICAL REVIEW. 


which are the hues that deck the Hours in 
successively changing tints. 
Carcinus fr. 8. 
év Spa povov tduov (al. ndiov) dv Tovet POdvos. 
Aurel yap adTd T TO KTHa TOS KEKTNLEVOUS. 
Porson corrected id:ov to dixkavov. It might 
be, I think, orovdatov. Then, Aurel yap ody 
TO KTH. ‘ 
Python /. 1. 
éoriy 8 Srrov pev 6 KdAapos Tépvy’ de 
T hérwp caopvov. 
Possibly arépwy’? ‘a wing—but not of 
birds,’ i.e. a building. 
Sosiphanes. fv. I. 
payous éxwodais Taca Mecoadj{s Kopy 
Wevdys ceAnvys aifépos KaratPares. 
Perhaps Wevder wedyvnv aifépos katarBarw. 
Sositheus 2. 
TH pa 0 év ymepa. 
Sawval r éurns cvvtiOnow eis TEXOs. 
Possibly Sawis ér dumvy ocvvtibyo’ olvov 
ydvos. 
Adespot. 458. 
det 8 dpavri 7’ 6&0 Kal tupdos jv. 
This should be, I imagine, 
del 8 dpav tis 6€b Kal TuPASds TEP Tv. 
‘He (Oedipus) ever had a kind of sharp 
sight even in his blindness.’ 
Ropinson ELLs. 


NOTES ON THE [lodireia *“AOnvaiwv. 


15 § 1. pera d€ tratdta ds e&érece 70 Sevtepov 
gre. pdduora EBddpm peta THY KAB0d0v —od yap 
mov xpovov KaTELXeEV. 

Wilamowitz-Moellendorff is unquestion- 
ably right in his judgment that the text is 
corrupt. The statement is that the duration 
of the second dpyxy of Pisistratus was six 
years—a year longer than the first dpxy, 
and little or not at all shorter than, on this 
supposition, the third dpy7 must have been 
[17-(6+5)= 6]. If so, ot yap odtv 
xpovev is simply nonsense. Wilamowitz 
hesitatingly proposes tpitw ; but the change 
of the numeral is, as he confesses, arbitrary, 
and there is no palaeographical motive. 
Besides, even two years is too long. Hero- 
dotus in his account (i. 61) gives no definite 
note of time, but the whole impression of 
his narrative is that the retirement of 
Pisistratus followed hard upon his restora- 
tion, certainly well within the space of a 
year. And the phrase od yap roAtbv xpdvov 
leads us to expect here a statement con- 
firming the impression we had _ before 
received from the dateless record of 


Herodotus. We expect to read: ‘In ‘the 
twelfth year after he first seized the tyranny, 
he was restored...Then when he went into 
exile a second time, in the twelfth year after 
he seized the tyranny ’—‘ Oh, but ’—objects 
the reader—‘ that was the year in which he 
was restored.’ ‘Certainly,’ replies Aristotle, 
‘but he was only in power a short time.’ 
That is what ydp (in the light of the 
Herodotean passage) seems to imply. 

And so, I believe, Aristotle wrote, though 
he expressed it somewhat differently. The 
difficulty in the text is due to the presence 
of an explanatory interpolation. We meet 
in the case of the immediately preceding 
date (14 § 4) a clear instance of such inter- 
polation. There ascribe, not understanding 
that gre SwSexdrw referred to the starting- 
point of the zpwrn katdoracis, introduced 
the erroneously explicit pera tatra and so 
wrought confusion in the sense. So here. 
Aristotle wrote: ‘when he was expelled for 
the second time in the seventh year (after 
his first ewile)—(the year of his restoration 4 
—certainly) for he held the power for only 


* 





THE CLASSICAL REVIEW. 


a short time.’ The copyist, not seeing that 
the ‘seventh year’ was the same as the 
preceding ‘twelfth year,’ reckoned from a 
different starting-point, introduced pera Tv 
kaGodov to render explicit what was not 
quite clearly stated. So we shouid read: 


c , \ , 
ds é€éreve TO devrepov ere: wdiora Bd, 
r) ‘ ‘ , lal + 
—ov yap ToAdvtv xpovov KQTELXEV. 


Thus we obtain the following dates: 
First dpyy: 5 years (561/0—556/5). 


Second dpyy: part of a year (550/49). 
Third dpyxy: 12 years (540/39—528/7). 





First exile: 6 years (556/5—550/49). 
Second exile: 10 years (550/49—540/39), 





Total : 


33 and part of a year (561/0 
—528/7). 


It is easily seen that, on this plan of 
reckoning, the total period of dpx7 might be 
variously reckoned at (a) seventeen and a 


_ fraction or roughly seventeen, and (0), by 


counting the first and the third dpy7 as hay- 
ing lasted each for some months beyond the 
round number of years, at nineteen. (qa) is 
represented by the passage in the Politics, 
viii. 13506; and also results if we subtract 
the sixteen years of exile from thirty-three ; 
(6) is adopted in 17 § 1. 

22 § 2. apGrov pev ody eran TeuTTH pero 
TavtTnv THv KatdoTtacw ed ‘Eppoxpéovtos 
dpxovtos TH Bovdy Tots TevtaKxociois TOV OpKoV 
erolnoav oe ereita. TOUS oTpaTyyous npovvTo... 
ere. O€ pera TaiTa SwoeKdTH ViKHTaYTES TV EV 
Mapadave paynv x.t.X. 

There is not necessarily a chronological 
mistake here. The date of Hermocreon’s 
archonship (504—503) only concerns the 
introduction of the Oath for the Five 
Hundred. Subsequently the ordinance for 
the election of the strategoi was passed, 
namely in 501—500, the twelfth year 
before Marathon (490—489). The passage 
is usually interpreted as if it were erev be 
TépTTw...dpxovTos mpatov pev TH Bovdy... 
éreta ~8é...; which would imply an in- 
consistency.1 The only difficulty is the 
discrepancy between Aristotle and Diony- 
sius, who gives Akestorides as the archon 
of 504-3 (v. 37). ‘ 

26 § 1. xara yap Tovs KaLpOUS TOUTOUS ove: 
mere pnd wyeunova exew Tors EmLEetKETTEpOUS 


1 On that supposition the best correction would be 
(not to stir réumtw but) to alter éme:ra to émeira <6 
éret 4>—the eighth year uera TavTHnY THY KaTa- 
TTATW 


107 


XN aard . : 

GAA auTwY TpoeaTtavat Kiuwva roy M 
, J 4 ‘ ‘ 7 

vewtepovt dvta Kai T™pos THY ToAWw owe 

Govra. 


t(ATiaoor 


TpomeA 


For the genuineness of the suspicious 
word Mr. Walker (Class. Rev. vi. 98) has 
made the best case that can be made, main- 
taining that in estimating the chronological 
difficulty we must not go by the received 
dates of this period, but must first draw 
the conclusions which are implied by the 
supposed presence of Themistocles at Athens 
in 462. But even if we granted—what we 
need not grant—that Aristotle altered his 
chronological view of the period so as to 
harmonize with this anecdote, we should not 
get rid of the objections. For there is 
something very suspicious, as Mr. Sandys 
rightly pointed out, in the combination 
VEWTEPOV Kal TpOs Ti TOAW dWe tporedOdvTa. 
Wilamowitz-Moellendorff hardly goes too 
far in calling it ‘unsinn.’ He supports the 
emendation vw6pdrepov with great ability 
(Aristoteles und Athen ii. 136), though he 
admits that it is ‘ein grobes Wort und von 
dem Euphemismus der attischen Eleganz 
weit entfernt.’ The reference to Rhetoric ii. 
1390b will persuade many; and perhaps 
rightly. But the correction is not so certain 
that it may not be worth while to put forward 
another conjecture which is palaeograpbi- 
cally easier and gives equally good sense. 
The author evidently means to say that 
Cimon had no political shrewdness or tact, 
and this natural lack was not in any mea- 
sure compensated for by sheer dint of 
political experience. He was devoid of that 
ctveots Which marked, for instance, Thera- 
menes. I therefore propose : 


MIATIAAOYACYNETWTEPON 


i.e. Kipwva tov MAriadov douveTwTepov OvTa. 

The resemblance of the tirst letters of the 
word ACY to the last letters of the foregoing 
word, AOY, misled the scribe into omitting 
them; and the surviving vow nihili verer 
repov could become nothing but vewrepov. 

30 $ 3,4. wai eis enavrov BovAcvew O€ 7 
div doxy adrots dpirra éfew x.7.A. 

It is recognized that a sentence ends at 
Bovdevey, and that an infinitive has fallen 
out before dé. BovAcerOa is adopted by 
Blass and Mr. Sandys, but does not account 
for its own omission. Read : 

Bovreverv. <KeAeverv> 82 & dy «7A. 
The change of & to 7 was a consequence of 
the omission of xeAever. 


108 THE CLASSICAL REVIEW. 


33 ad fin. rodeuov te Kabeataros Kal ék 
TOV OTAWY THS ToOATELas OvONS. 

It is remarkable that the particle ye does 
not occur in the ’A@yvaiwv TModrteia. Is it 
possible that we may have it here? zoAé€nov 
ye xafeoratos would be an improvement 
(‘considering that there was war’), and 
would suggest the sense desired by Her- 
werden who proposed KalTrep ToAE MOV. 

35 § 2. Kai Tovs T EgudArou kat’ ’ Apye- 
OTPATOV vopous TOUS epi TOV ‘Apeomayirov 
KabetAov ee *Apevov mayov, Kal TOV SoAwvos 
Geopav dorou StappiaBytHoes eixyov Kal TO 
Kdpos ry nV év TOUS dukaorats KkareAvoav os 
exavopboivres Kal Towodvtes avaudisByrntrov 
THY ToAurelav. 

Two distinct attacks were made on the 
Areopagus, the first by Ephialtes, the second 
by Pericles. There is now. no justifica- 
tion for Sauppe’s rejection of the words 
kat IepuxAjs in Arist. Pol. ii, 12 (1274a): 
Kal THv pev ev Apetw tayo Bovdnv ’EdudArys 
éxoAovoe Kat IlepixAns. The notice in our 
treatise, 27 § 1, rév “Apeotayitav évia. 
mapetAero saves, beyond dispute, the suspected 
words. But it is to be observed how the 
misinterpretation of such a phrase led to 
the false notion, found in Plutarch, that 
Pericles was the real mover in the attack 
which was headed by Ephialtes. The reform 
of Ephialtes was previous to the prominent 
appearance of Pericles on the political 
stage; but it is not unlikely that when 
Pericles resolved to ‘dock’ some of the few 
privileges which Ephialtes had left to the 
Areopagites, he did not make the motion 
himself but got another to make it for 
him. This is the only supposition on 
which we can explain the words in 35 § 2. 
What, we are entitled to ask, did they 
leave unstirred the well-known laws in 
which Pericles carried on the policy begun 


by Eph‘altes, and only remove those of the 
obscure Archestratus of whom or of whose 
laws we never hear elsewhere? The con- 
clusion seems unavoidable that Archestratus 
was the instrument of Pericles, and that he 
is here mentioned because the laws were in 
his name. 

But the sentence requires a slight emen- 
dation. According to the text of the 
papyrus tovs 7’ corresponds to cal rév 
Sodwvos «.7.X. But it is clear that xabetdov 
refers only to the Areopagitic laws, not to 
the Solonian which were preserved in the 
Prytaneum ; and rév SoAwvos Oeopav cannot 
be separated from xai 76 xdpos, both depend- 
ing on xatéd\voay. It has been suggested 
that 7’ should be omitted : but why should 
it have been inserted? The true restora- 
tion is 


Kal <tovs> ’Apxeorparov vopous 


47 § 5. clo péperan fev ovv eis THY BovAnv 
Ta ypappar( eta Ta] Tas KataBords avayeypa.- 

éva. 

So Kaibel and Wilamowitz. But x stands 
before ras in the papyrus. Read perhaps : 


A / 
TH ypappareta TH TevTEeKaideKa. 


1€ (as often IC) might have become K. 
Fifteen ypaupareta have been mentioned : 
ten containing the instalments paid in each 
prytany ; thre ee for those paid three times a 
year; one for those paid in the ninth 
prytany; and one, in the case of the 
Basileus, specially reserved for the revenue 
arising from reméevn (ev ypappatetors AedevKw- 
pevors does not mean that he used more than 


one at a time), pies 
5 Sy OUR. 


ON TIBULLUS I. 1, 2. 


et teneat culti iugera multa soli. 


In the Classical Review, May, 1894, p. 
198, the reading magna is defended by F. 
K. Ball, with comparison especially of Ov. 
Amor. 3, 15, 12, and Statius, Zhed. 5, 550. 

1. The MSS. read as follows: magna, 
AVg; mulia, G, Par. Fris. Diomedes (the 
MS. testimony is reversed in the note 
above mentioned) ; and the combination of 
both the excerpts with Diomedes is here on 
the whole to be preferred to AV, 


2. Iugera occurs but four times in the 
Tibullus collection, as follows : (a) i142 
(this passage). (0) ii. 3,42: multa innwmera 
dugera. (c) iii, 3, 5: multa iugera. (d) 
i, 3, 75: novem per cugera. In these other 
cases the text is not in dispute, save that 
Baehrens characteristically wishes to change 
the multa at ii. 3,42 to culta! Multa is 
therefore rather strikingly in harmony with 
the usage of Tibullus. 

3. Similar are: Ov. Fast. iii. 192: cugera 
pauca soli ; ex Pont, iv. 9, 86: tugera multa 


a 





THE CLASSICAL REVLEW. 


rett; Verg. Georg. iv. 127: pauca relicti 
ugera ruris ; Juv. 9, 60: iugeribus paucis. 

4, Multa is quite in harmony also with 
the spirit of this first elegy, in which the 
poet so often suggests that, while he once 
possessed many acres, he now has but /ew ; 
cf. vv. 5, 19—20, 37, 41. 

5. The whole passage in Ovid (Amor. 3, 
15, 11—14) reads : 

Atque aliquis spectans hospes Sulmonis aquosi 
moenia quae campt iugera pauca tenent, 
quae tantum’ dicat ‘ potuistis ferre poetam, 

quantulacumque estis, vos ego magna voco. 

(a) The two most important Ovid MSS. 
(the two Paris MSS.) are not available on 
this passage. 

(6) To back up parva here, as K. P. 
Schulze does, by referring to Hor. Sat. 1, 6, 
7 (Olim qui magnis legionibus imperitarent) 
is idle. For (1) legionibus here refers to 
troops of the Etruscans, who had no Roman 
‘legiones’ ; and (2) legionibus here naturally 
= exercitibus, a usage sanctioned by the 
highest authorities of the Augustan age (as 
well as other periods), e.g. Verg. Aen. 9, 
368: cetera dum legio campis instructa 
tenetur ; Aen. 8, 605: de colle videri poterat 
legio; Xe. 

(c) Parva would seem too strong an ex- 
pression here; for while pauca might be 
accepted as poetic hyperbole, parva would 
appear to carry with it an almost contemp- 
tuous seriousness which Ovid would scarcely 
apply to Sulmo, especially since it is far 


109 


from true that the plain of Sulmo was a 
narrow one. 

(d) The perva might easily have been 
interpolated by a misunderstanding of the 
passage arising out of a confusion of iugera 
with moenia, in accordance with which the 
interpolator might have thought he was 
properly contrasting parva with the magna 
of v. 14. 

6. The passage in Statius, 7heb. 5, 550 
(spatiosaque iugera complet) is an outrageous 
hyperbole, as it refers to the size of the 
dragon that had just killed the child 
Archemorus. It may be explained as a 
hypallage for spatia iugeralia; i.e. the 
dragon is said to be stretched out over a 
space as big as acres. It does not therefore 
seem a fair parallel to the serious expres- 
sions magna vugera and parva iugera. 

7. The ‘magna’ in our Tibullus passage 
may have arisen from some copyist’s over- 
sensitive ear to assonance, iugera magna 
being easier to write from dictation than 
cugera multa. 

8. It seems doubtful, therefore, whether 
any Latin parallel to the English idiom, 
‘broad acres,’ exists. If one had been in 
vogue, would ‘magna’ have been the 
adjective? or something like patentia or 
diffusa ? 


Kari P. HARRINGTON, 


University of North Carolina. 


SUETONIUS, VERO, 45. 


Alterius jstatuae| collo ascopera deli- 
gata simulque titulus: Ego quid potui! 
sed tu culleum meruisti. 

THE above is the reading of all the 
editions. Ascopera is a conjecture of 
Poliziano, for which ascopa is found in all 
MSS. The passage describes an incident 
at the end of Nero’s reign, when derisive 
mottoes were attached to his statues or 
written up elsewhere. ‘The words eyo quid 
potut have puzzled the editors. One inter- 
pretation is ‘Ego quid potui (sc. peccare) ?’ 
Baumgarten-Crusius in a long note says 
that the ascopera attached to the neck of 
the statue signified, ‘abeundum iam esse 1n 
exilium et miseriam Neroni,’ and he goes on 
to explain the words put into the mouth 
of the ascopera; ‘Ego cui nulla potestas 
in civitate sed iustum de te iudicium quid 


potui nisi hoc % sed tu, si leges valent, ut 
matricida culeum meruisti.’ 
Neither of these interpretations carries 
conviction. The latter evidently adopts the 
gloss of Suidas: ’Ackomypa* TO paprurioy 
nrow TO caxkoTvOviov (wallet, knapsack) and 
throws emphasis on the latter part of the 
compound—-ypa. There appears to be bo 
authority for this meaning except Suidas. 
A similar gloss is found in Ducange (Gloss. 
m. et t. Lat.) s.v. ASCOPERA idem quod 
ascopa, marsupium. But he gives a que 
tation from a Chronicle of the tenth cen- 
tury: Cum sciret non nisi in ascopera mis! 
modicae quantitatis vinum haber, where 
a. obviously means doxos, uter. And Lewis 
and Short quote from the Vulgate, Judith 
10, 5: Imposuit itaque abrae suae asco- 
perai (v./. ascopam, LXX. doxorv@mor) vin. 


110 


This is satisfactory evidence that in later 
Latin ascopera was used of a receptacle for 
liquids like the simple doxos. If the usage 
may be assumed to reach back to Suetonius’ 
time we can get a much more pointed 
meaning out of the words by punctuating : 
Ego quid? potui ; tu autem cullewm meruisti— 


THE CLASSICAL REVIEW. 


where potui is the dative expressing the 
purpose, ‘the predicative dative,’ and the 
meaning is: ‘What am I? A sack to drink 
from. But you have deserved a sack of 
another kind.’ The allusion is of course to 
the well-known punishment of matricides. 
W. CHAWNER. 





NOTE ON HOR. 


Iv may seem rash to attempt to emend 
an author like Horace, who is so much 
read, and has had so many editors ; and it 
may be thought presumptuous to meddle 
with a passage on which so many well- 
known scholars have spent their efforts ; 
still the emendation which I have to propose 
is at once so simple and so probable that 
I can think of no reason why it should 
have escaped so many generations of editors. 

The traditional reading of the best MSS. 
in this passage is :— 


teque dum procedit, io triumphe ! 
non semel dicemus, io triumphe ! 


and this Hirschfelder (4th ed. of Orelli) 
retains, finding none of the emendations yet 
suggested an improvement on the MSS. 
With this reading Triumphus is personi- 
fied (as it is taken in Hpodes ix. 21) and 
procedit refers to Augustus. But it should 
be noted that throughout the poem the 
second person refers to Iulus as in verses 2, 
33, 41, and again in v. 53, the first line of 
the following stanza. The variant procedis 
does not eliminate the difficulty of the 
passage, whether the 2nd person is under- 


OD. IV. 1. 49. 


stood to refer to Augustus or to the 
personified Triumphus. 

‘Procedit’ should clearly be kept, from 
MSS. evidence as well as from intrinsic 
probability ; and the word can only refer 
to the ‘triumphator.’ The next step is 
to get rid of the pronoun: editors who 
have done this are Meineke and Bentley ; 
but the former’s ‘atque’ is as prosaic as the 
latter’s ‘isque’ is weak and un-Horatian : 
nor can one see how the reading ‘te,’ which 
has the consensus of the MSS., could have 
arisen from either of them. 

I propose to read ‘¢erque’: ‘and we, as 
Caesar doth advance, the citizens, not once 
alone will raise the cry “Jo Triumphe!’ 
but thrice “Io triumphe!”’’ ete. 

I do not wish to lay any stress on the 
antithesis ‘ter...non semel’ (the Greek 
ToAAdKis Te KovX amas) but rather on the 
‘thrice-repeated cry’ io triumphe, with 
which one may compare the threefold repeti- 
tions in the chant of the Arval Brethren, 
also the thrice-uttered Greek shout of 
triumph ‘rjveAda Kaddirixe’—KadXivixos 6 
tpitddos Kexades, Pindar Ol. ix. 2. 


BepiStov. 


In Julian’s charming epistle to Evagrius 
(numbered 46) occurs a word Gepisiov, which 
ought to find a place in the lexicons :— 

TOTO wot pepakiw Kopion véew Oeptdvov 
éddxer pidtartov. 

Oepidiov is formed from Gepi~w as XEuLdOLov 
from yepalo. 


L. and §. quote Xenophon and Aristotle 
for Gepi€o with the meaning of passing the 
summer; but omit the noun. The whole 
of this letter of Julian shows that he uses 
it here to indicate a dwelling for the 
summer. 

E. J. CHINNOCK. 








THE CLASSICAL REVIEW. 


SUMMERS ON THE ARGONAUTICA. 


A Study of the Argonautica of Valerius 
Flaccus, by Water C. Summers, B.A., 
Fellow of St. John’s College, Cambridge. 
(Cambridge: Deighton, Bell & Co. 
don: G. Bell & Sons.) 2s. 6d. 


Lon- 


A warm welcome must be given to this 
little book, which forms an excellent intro- 
duction to the study of a poet not much 
read in this country. It is divided into nine 
sections in which the following subjects are 
discussed: the name of the author and 
scope of the poem, similarities of thought 
and language between Valerius and later 
writers, the indebtedness of Valerius to his 
predecessors, especially Apollonius Rhodius 
and Vergil, the syntax and metre of Valerius, 
and finally we have some literary and textual 
criticism. Mr. Summers has treated his 
subject thoroughly and systematically from 
his own point of view, but it may perhaps 
be regretted that he has not made it a more 
complete ‘study’ in itself by incorporating 
the results of the investigations of Schenk, 
Thilo and other scholars who are often 
quoted. As it is, the work of Mr. Summers 
has to be supplemented by continual 
reference to their writings, which are more 
or less difficult of access. Thus the date and 
life of Valerius are not discussed because 
‘they have been adequately treated by other 
writers, especially Thilo.’ We should like 
to see a short statement of Thilo’s results. 
Again, in giving lists of contradictions and 
lack of connexion in the Argonautica, and 
of imitations from or by other writers, Mr. 
Summers only makes additions to the lists 
given by other scholars. I can only say I 
think it a pity that he has so limited 
himself. 

There is little doubt that the poem, like 
most other Latin epics, was intended to be 
in twelve books, and the only question is 
whether its unfinished state—for we have 
only about seven and a half books—is due 
to Valerius himself or to some other cause, 
The argument of Baehrens to prove that 
Valerius lived to finish his work amounts to 
no more than the statement that from a 
calculation of dates he had time to do so, 
and that it was customary (1) with the Epic 
poets of Rome to write and recite a book 
each year—assuredly a debile fundamentwm 
(as Baehrens himself allows) on which to 

build his conclusion! On the other hand, 
the absence of reference in Statius to any 


incidents on the return voyage of the Argo 
does not go far to show that Valerius did 
not finish the poem (the view to which Mr. 
Summers inclines) when the admitted rarity 
of reference to these later incidents in other 
writers is taken into consideration. Schenkl's 
alleged proof of the original incompleteness 
of the Argonautica from examples of con 
tradictions and want of connexion in the 
text at various points may be met by the 
supposition that Valerius, like Vergil, lived 
to complete his poem but not to revise it, 
for the examples given by Schenk] and added 
to by Mr. Summers are not more remarkable 
than similar examples in the Aeneid. 

However it be, this point will probably 
remain a matter of conjecture, as will also 
the question how far, in his account of the 
return voyage, Valerius diverged, or intended 
to diverge, from Apollonius. Certainly the 
ingenious parallels that are brought forward 
from the Orphic Argonautica tending to 
show that the author of the latter had 
Valerius before him—a relationship which 
Mr. Summers claims, and apparently 
with good reason, to be the first to have 
expounded—point to a great divergence 
from the Greek. This imitation, however, 
if once admitted, is also evidence pro 
tanto that Valerius did complete his work. 
Consequently, to save the opposite theory, 
Mr. Summers has recourse to the highly 
improbable hypothesis that the account of 
the return voyage in the Orphic poem * may 
be due to some plan of it left behind by 
Valerius.’ 

Mr. Summers acutely traces the debts 
owed by Valerius to his predecessors as 
regards the subject-matter, but 1 cannot 
agree that there is much likelihood of 
Valerius having had the present Scholia of 
Apollonius before him, without further 
evidence that these Scholia were in existence 
at that time. The three Scholiasts on 
Apollonius whose names are given are 
Lucillus of Tarrha, Sophocles, and Theon. 
Of Sophocles nothing is known, of Lucillus 
hardly anything, and if Theon, as 1s thought 
likely, was the well-known grammarian of 
that name and father of Hypatia, his date 
is of course much later than Valerius. No 
doubt an unmistakable reference of Valerius 
to these Scholia would be a different matter, 
and would in its turn help to date them for 
us, but the two references given are merely 
conjectural. Far the most important pre- 


112 THE CLASSICAL REVIEW. 


decessors of Valerius are Apollonius and 
Vergil, and to them naturally most space is 
devoted. The influence of Apollonius is 
both direct and indirect. The direct influence 
appears under three aspects : (1) more or less 
literal renderings of the Greek, (2) similes, 
(3) various episodes. Of the episodes it is 
said that Valerius ‘ likes to treat with com- 
parative brevity points which had already 
been dwelt upon in detail, and conversely.’ 
From this I see no reason to dissent, but 
when it is added as an example ‘that the 
gathering of the heroes which takes up only 
a score or so of lines in Greek, occupies here 
[in Valerius] a couple of hundred,’ I am not 
sure that I understand what is meant. In 
Apollonius the gathering of the heroes, 
which I should take to mean the catalogue 
of the Argonauts, occupies 211 lines, not 
merely a score or so. The indirect influence 
of Apollonius is found all through the poem. 
Most readers will agree with Mr. Summers 
that Valerius is superior in artistic arrange- 
ment and probability. Two pages are 
devoted to acomparison between Apollonius 
and Valerius in their treatment of love. 
On this well-worn theme I do not desire to 
say much, because after all is said each 
reader will probably retain his own view. 
I do not however consider that Prof. Ellis 
in his review of the present work in the 
Academy some time ago has quite done 
justice to Mr. Summers on this point. He 
writes: ‘I cannot agree with Mr. Summers 
in his apparent preference for Valerius in 
his treatment of love: he finds this 
superiority in the gradual and artistic 
development of Medea’s passion. But the 
natural frigidity (which he admits) of the 
Roman always makes itself felt, nor can it 
be said that his genius led him instinctively 
to the exhibition of female passion. Whereas 
from the moment when Apollonius’ Medea 
appears upon the scene, it is perceptible that 
the poet has reached the point of real 
interest, the vital centre of his art. Tull 
then he is the mere narrator ; thenceforward 
he is identified with his heroine, and steps, 
so to speak, on the stage in his own person.’ 
With the excellent judgment herein con- 
tained I agree emphatically: at the same 
time Mr. Summers hardly goes so far as to 
pronounce in favour of Valerius in his 
description of love on the whole. What he 
says is, ‘The strong point in the description 
of Valerius is that he has depicted the 
gradual growth of love better than either 
of his predecessors.’ Upon this however I 
should at once join issue, Vergil’s Dido I 
put aside, as a delineation incomparably 


superior to anything in Apollonius, much 
more in Valerius ; but, after all, is there 
such a thing in ancient classical literature 
as the gradual growth of love? I fail to 
see it either in Euripides, Apollonius, 
Vergil, Ovid, or Valerius. Love is repre- 
sented as something introduced into the 
human soul from outside and is entirely 
beyond the control of the subject of the 
passion. It is a fate, a disease, a heaven- 
sent plague. Hence the mechanical con- 
trivances for its production which are so 
tasteless to us, such as the arrows of 
Eros &e. In Apollonius Eros shoots an 
arrow at Medea. It takes effect and inflames 
her with the love for Jason required for the 
purposes of Hera. Cupid takes the form 
of Iulus noctem non amplius unam and in- 
spires Dido with love for Aeneas. Ovid’s 
heroines are victims to fate, Phaedra, 
Ariadne, Hypsipyle. Medea says of her love 
for Jason et me mea fata trahebant. Is it 
different with Valerius? We read that 
Juno borrows the cestus of Venus and 
repairs to Medea in the guise of her sister 
Chalciope. Mr. Summers says ‘the use of 
the cestus is not at all clear.’ Butsurely it 
is very clear, its use is the same as in the 
14th book of the Jad, viz. to give the 
wearer the power of inspiring love whether 
it be for herself, as in Homer, or for some 
one else, as in Valerius. It is true that 
Medea had had a previous chance meeting 
with Jason, but that was not enough. It 
was necessary that she should have an all- 
absorbing passion for him and this had to 
be brought about by the application of some 
external force. What appears to be the 
gradual growth of love is, I venture to 
think, rather the various manifestations of 
the passion which, subsisting there all the 
time, a full-blown rose from the first, is 
ready to be called into action at the 
‘psychological moment.’ If these remarks 
are just, and they enunciate the same view 
as that taken by Sainte-Beuve in his well- 
known article on the Medea of Apollonius, 
then the alleged superiority of the Roman 
poet falls to the ground. 

The sections on the imitations from 
previous Latin writers (in which Mr. 
Summers refutes Baehrens, who denies any 
imitation from Ovid and allows very few 
from Lucan), on the syntax and the prosody 
of Valerius are very well done. I cannot 
help thinking that in one point, where it is 
said that of the genuine trochaic caesura we 
have four or five examples while Mueller 
admits only one, Mr. Summers has misunder- 
stood Mueller, who seems to have meant only 


~~ eed 















faye o 


> 


deities 


THE CLASSICAL REVIEW, 


one case of a trochaic caesura without a 
hephthemimeral caesura too, Mueller’s 
example is iii. 191 


frater Hagen Thapsumque securigerumque 
Nealcen. 


In all the other cases cited by Mr, Summers 
there is a hephthemimeral caesura as well, 
It is clear that egé in viii, 158 cannot stand 
and we must adopt Mueller’s ego o or make 
some greater change. Speaking of Valerius’ 
style, Mr. Summers finds, peculiar to 
Valerius, ‘a fulness and copiousness of 
expression which adds little to the meaning.’ 
But surely this repetition of the thought is 
a marked feature of Vergil’s style too, to 
which Henry in his voluminous and amusing 
commentary constantly calls attention, 
speaking of it as theme and variation. 

The last section deals with the text in a 
judicious and conservative manner. ii. 641 


113 


cannot, | believe, be satisfactorily translated 


as it stands, nor does Mr. Summers make 
much of it. In vii. 156 we should surely 
read pudore with Baehrens (after Voss) for 
pudori, and thus at once restore sense and 
grammar. I have noticed a few examples of 
slight errors or misprints. At the bottom 
of p. 49 two more commas de needed to 
make the sense obvious. ‘I'wice the late 


Prof. Sellar has his name spelt Sellars, in 
the note on p. 21 Mén should be Mene or 
Méné, and on p. 71 westu is a misprint for 
ueste. It has naturally occurred to me to 
attempt to point out what I consider to be 
defects rather than to take up room with 
commendation, the former being a more 
useful if less gracious task, but I cannot 
conclude without expressing once more my 
sense of tho great merit of Mr. Summers’ 
work, 


R, C, Seaton, 


THE NEW EDITION OF PAULY’S ENCYCLOPAEDIA. 


Paulys Real-Encyclopidie der classischen 
Alterthumswissenschaft, neue Bearbei- 
tung, herausgegeben von G. Wissowa, 
(Metzler) Stuttgart. 1893—4.. Vol. I 
2902 columns (Aal to Apollokrates). 
30 Mk. 


Pauty’s Real-Encyclopédie, originally pub- 
lished in six volumes (1839—52), takes its 
title from its first editor, Auacust PauLy 
(1796—1845), who was one of the staff of 
teachers at the Gymnasium at Stuttgart. 


The first volume of the original work 


appeared in 1839, and, after the death of 


Pauly, the last three volumes were edited 


by Teuffel (1820—78) and Walz (1802— 
1857), the former of whom completely 
recast the first volume for its second issue 
in 1864—6. A new edition of the whole 


work is now in preparation under the 


general editorship of Dr. Georg Wissowa, 
Professor of Classical Philology at the 
University of Marburg, who is already 
known as the editor of the second edition 
of Mommsen’s Handbuch der rimischen 
Alterthiimer. He has secured the co-opera- 
tion of nearly 120 experts in different 
departments of Classical learning; the 
work will be comprised in ten large volumes 
of about 1450 pages each, and will be 
completed in ten years. 

The first volume, which has been published 
NO. LEXVI, VOL. IX, 


in two parts, now lies before us. It is 
practically an entirely new work, and, 
owing to its thoroughness and completeness, 
deserves the warmest welcome from all who 
know the value of a comprehensive and 
absolutely trustworthy book of reference 
in the departments of Classical Mythology, 
Geography, Biography, History, Literature, 
Archaeology, Art and Antiquities. One of 
the many advantages of the new edition is 
that it includes all the names of persons of 
any historical importance whatsoever. Thus, 
under the heading of Alexandros we have 
no less than 107 persons of that name ; 
under that of Annius, as many as 127 
members of the gens Annia, including Milo 
and Marcus Aurelius. It does not profess 
to compete with what has been justly 
described by Mr. Tozer as ‘that model of 
compendious learning,’ Pape -Benseler’s 
Werterbuch der griechischen Eigennamen, 
which necessarily includes many names 
entirely unknown to fame. Thus Agamestor 
son of Laius (Apoll. Rhod. 2, 852), Agasi- 
cles of Sicyon (Pausan. 2, 10, 3), and 
Agathophanes of Cythnos (known by an 
inscription alone) are to be found in Pape- 
Benseler, but not in the new Pauly. Simi 
larly, the Amaryllis of Theocritus, who 
lives in the Pastoralia of Longus, and is 
immortalized by Virgil and Milton, has 


apparently too shadowy an —- to find 


114 


a place in a Real-Encylopidie. On the other 
hand, while only six of the name of Agath- 
archus are recorded in Pape-Benseler, as 
many as fifteen are distinguished in Pauly. 

Among the most important articles in 
the present volume are Achaia (by Toepfter) 
with the most comprehensive account which 
has yet appeared of the constitution of 
the Achaean League; Aetolia (Wilcken), 
Aegina (Hirschfeld), Amphictyonia (Cauer) ; 
Aeschines, Andocides and Antiphon (Thal- 
heim); Alcibiades (Toepffer), Antiochus 
(Wilcken and others); Aeschylus and 
Alcaeus (Dieterich), Anthologia (L. Schmidt), 
Apollodorus, the Greek ‘grammarian’ 
(thirty-one columns, by Schwartz) ; Annaeus, 
including Lucan and the Senecas (by 
several writers) ; Alphabet (Szanto and Joh. 
Schmidt); Amazons (Graef), Altar (fifty 
columns, by Reisch) and Aphrodite (fifty- 
eight, by Diimmler). Agriculture and 
Botany are well represented by an im- 
portant article on Ackerbau, and by shorter 
notices of Apfel and Anemone (by Olck) ; 
Superstition, folk-lore c&e.. by Aberglaube 
(sixty-four columns, by Riess) ; and Greek 
Constitutional and legal antiquities are dealt 
with in numerous short articles (by Szanto 
and Thalheim). As an indication of the 
completeness of the work it may be noticd 
that the recondite question of the dpvyets 
BiBro in the Alexandrian Library, though 
mentioned by Tzetzes alone, is discussed in 
a column and a half, while even the 
Anonymi have eleven columns assigned to 
them. The articles in general, so far as I 
have consulted them, are written in a terse 
and clear, perhaps rather dry and decidedly 
‘objective’ style, and are equipped with an 
abundance of references to ancient and 
modern authorities. 

It is easy to find omissions in so large a 
range of references. Thus, in the articles 
on Antiphon and Andocides, one might have 
looked for, some mention of M. Georges 
Perrot’s 7’ Eloquence Politique et Judiciare a 
Athénes (les Précurseurs de Démosthéne), and 
also of Professor Jebb’s Attic Orators from 
Antiphon to Isaeus. In the article on 
Alkidamas, a reference might have been 
added to Mr. Cope’s notice of that rhetori- 
cian in his articles on the ‘Sophistical 
Rhetoric’ (Journal of Cl. and Sacred Philo- 


THE CLASSICAL REVIEW. 


logy, iii 263—8). In the article on dvdxeiov 
the statement that, in Aristotle’s ’A6. wrod. 
15 § 4, the reading is ‘really’ & To 
’"Avaketw is inconsistent with the fact 
that Mr. Kenyon has since withdrawn 
that reading in favour of év 76 Oncelw, 
and has been followed in this by all sub- 
sequent editors, even Wilamowitz, who 
was once most eager for retaining ’Avaxetw, 
having now acquiesced in Onceiw (Aristot. u. 
Athen, i 266, note 17). Under Anacreon, 
[Plato’s] Hipparchus, which is quoted as an 
authority for the poet’s call to the court of 
the Peisistratidae, should now be supple- 
mented by a reference to ’A@. vod. 18 § 1. 
Lastly, under aéixiov we should expect some 
notice of Hyperides, in Demosthenem, col. 
24, 15. 

The work is very sparingly illustrated. 
The only cuts in this volume are a diagram 
explaining the use of the abacus, a map of 
the neighbourhood of Amphipolis, a small 
plan of Akragas which, though good for its 
size, is inferior to those published in 
Schubring’s monograph and in Freeman’s 
Sicily, a fairly large map of Alexandria, 
and lastly an excellent plan of the temple 
and precinct of Amphiarrus at Oropos. It 
is a matter of some regret that the cuts are 
so few ; amphora and altar are completely 
unillustrated ; and we fear that when, in 
the fulness of time, we reach the subject of 
Vases, we shall have to rely on other works 
of reference if we are to retain a vivid 
apprehension of all the variety of type that 
is characteristic of these interesting relics 
of antiquity, Let us hope that the pub- 
lisher may repent, and that, some ten years 
hence, the success of the present work will 
have proved to be so great as to prompt 
him to produce, as a thank-offering at its 
close, a handsome supplementary volume of 
illustrations only, executed in the excellent 
style that is not unknown at Stuttgart. 

Meanwhile, the work which has made so 
good a beginning fully deserves to find a 
place in the Library of every College in 
England and the United States, and also on 
the shelves of every scholar who desires to 
keep abreast with the latest results of 
modern research. 

J. E. Sanpys. 





THE CLASSICAL REVIEW. Bk 


A NEW THEORY OF WORD-FORMS. 


Die Entstehung der Dehnstufe. Von Witneth 
STREITBERG. Strassburg: Triibner. 1895, 


Ty the armoury of philologists for the last 
twenty years accent has been the most 
important weapon. A theory of accent 
produced Brugmann’s doctrine of sonant 
nasals and liquids, another theory of accent 
produced Fick’s explanation of the inter- 
change of e and o vowels in the same root. 
The new theory with which this paper deals 
is also founded upon accent, although it 
rises in the end to a suggestion that the 
physical theory of the Conservation of 
Energy extends to language. The theory 
deals with processes which took place in the 
original Indo-germanic language, but which 
have parallels in its modern descendants. 
Moreover it deals with words in their 
finished and complete form, As the author 
says with justice, the theory is not exposed 
to the objections which attach to all schemes 
of root-expansion and the like—theories 
which he obviously regards, as do others, 
with considerable suspicion. 

Ever since modern theories of sound- 
change began, the question has repeatedly 
been raised: Why, for example in Greek, 
should we have such forms as zatyjp with a 
long vowel while all other cases from the 
same stem matépa, ratpac1, matpos have 
either a short stem-vowel or no vowel at 
‘all? Why in Latin pés but pédem, pedis ? 
Again, why Zevs with an acute accent, vais 
with a circumflex? To all of these ques- 
tions various answers have been given at 
different times. But like the priest of Nemi 
each answer obtained sway by the slaughter 
of its predecessor and held its power only 
till a better came. And change followed 
change with wonderful rapidity. The present 
theory, if not armed against all conclusions, 
seems at least less easy of overthrow than 
its predecessors. 

The theory which Professor Streitberg 
sets forth in this treatise—a reprint from 
vol. iii. of the Indogermanische Forschungen— 
is notin all respects new. Several scholars, 
Johansson, Bechtel, Michels, have contri- 
buted among them great partof the materials, 
All that Professor Streitberg modestly claims 
to have done is to co-ordinate the results of 
the scholars mentioned, and to have cor- 
rected them where correction was necessary. 
_ In fact, out of scattered hints unsupported 
er but scantily supported on the part of 


their authors by evidence, he hag built up 
the structure of a complete theory. 
Curiously enough he seems not to know of 
a scholar who had promulgated an attempted 
scheme earlier than any of the three scholars 
to whom he refers. In the introduction to 
Den graeske Nominal-flexion published by 
Dr. Torp at Christiania in 1890 a eruder and 
less complete scheme than Dr, Streitberg’s 
is to be found. Whether, however, it is 
Dr. Torp’s own, or borrowed, there is, if I 
mistake not, nothing in the little book to 
show. 

The question to be answered is: What 
are the causes why original short vowels 
should be found lengthened in certain 
definite groups of instances? Dr. Streit- 
berg’s reply is: ‘If a mora has been lost in 
a word, an accented short syllable im- 
mediately preceding the lost mora is length- 
ened, while a long syllable immediately 
preceding, if it has the acute accent, changes 
it to the circumflex.’ 

The loss of a mora may take place in one 
or other of three ways: (1) a whole syllable 
may be lost, (2) a long syllable succeeding 
the accented syllable may be shortened, 
(3) a long diphthong like du, du, dn, dn (for 
uw and discharge in such combinations 
precisely similar functions) may lose its 
second component. Dr. Streitberg brings 
analogies from many modern languages to 
show that a similar rule prevails there also. 
An illustration from English, which he does 
not mention, will elucidate the rule as well 
as another. If the rule propounded above 
were to hold true in modern English, a 
disyllable like cd@nnot (pronounced cdnot) 
should be represented when reduced to a 
monosyllable by a syllable containing a long 
vowel cant, the two morae represented by 
the two short vowels being now represented 
by one long vowel (=two morae). This is 
really what happens, for the result of re- 
peated experiment is to show that can’t 
takes precisely the same time to pronounce 
in ordinary conversation as cannot. 

The so-called monosyllabic root-words are 
therefore a development out of disyllabic 
words ; the Dorie zs (the vocalism of the 
Attic zovs is still unexplained), the Tatin 
pes represent therefore a more orig 
*nodos or *pédos. The Latin vox must be 
carried back to a form *xyogos parallel to 
the Homeric accusative (*)éra, parallel also 


to the neuter word féros. Disyllabig and 
12 


116 


monosyllabic forms from the same root often 
stand side by side, naturally often with a 
difference of meaning: dopos, Pup; KAoz7ds, 
kop. The Greek 66 in yadxoBarés 66 and 
the like represents an older *dém parallel to 
ddpos and Lat. domus. It is impossible here 
to go into further detail on such stems. A 
word, however, on two of special interest : 
Zevs, Sanskrit dydus; Bots, Skt. . gdus. 
These stems are distinguished, according to 
Streitberg, from those which have an original 
long vowel by this point, amongst others, that 
in the latter class the long vowel is carried 
throughout the paradigm; vdaes, vies but 
Boes, Lat. navés but bdvés. The original 
forms, therefore, corresponding to Zevs and 
Bots were *diéuos and *géwos, from which 
came according to this law *diéus, *géus ; 
the original form for vads was *nayos. We 
now see that the substantive yAaté may 
owe its circumflex by the side of its adjective 
y\avkds to the loss of a mora in an original 
form *gléukos. The diphthong with acute 
accent represents two morae, the diphthong 
with the circumflex three. 

Another interesting series are the root- 
stems in composition. Streitberg shows 
that root-stems are commoner in composition 
than as separate words and that an .o-stem 
as a separate word often has a root-stem 
parallel to it in compounds ; cp. fvyév with 
vedtvé, Lat. semi-fer with ferus. Where the 
accent passes forward to the first element, 
there a so-called root-stem has arisen in the 
second element by the loss of final -os, and 
root-words like zrvé arise by resolution of 
compounds into their component parts, the 
separate word being properly of the type 
*rrvxos. The numerous Sanskrit root-stems 
in composition followed by -é- (g6-74-¢ ‘ cattle- 
winning,’ dharma-dhf-t ‘law-observing’ &c.) 
are successfully explained by Streitberg as 
reduced forms of -to- stems, stems which 
were not originally and never are exclusively 
passive. 

It is now clear how the question as to 
the origin of zarjp can be answered. It 
represents an earlier *patéros. The suffixes 
“pay (rounv), -pdv (ayendv), -js (edyevijs), 
-ws (yds) stand side by side with suffixes 
having a short stem vowel followed by -o-, as 
in the participial -evo-s &c. The -s-stems 
had no doubt at one time more parallels 
with a vowel suffix than Streitberg allows. 
The Homeric yeve} must represent *yeveor) 
by the side of yévos. These stems with a 
vowel suffix must surely have existed to a 
large extent in early Latin ; otherwise it is 
hard to see why from consonant stems we 
should have verbs of the type gener-a-re, 


THE CLASSICAL REVIEW. 


moder-d-re &e. With regard to forms of 
the type of im7eds also it is difficult to feel 
that Streitberg has reached a satisfactory 
conclusion. He would make them. arise 
by contamination between the type repre- 
sented by the Zend bdzdus (= original -duo-) 
and the type represented by the Sanskrit 
acvayus. But if *diéwos becomes Zevds why 
may not *ekuéwos become tr7evs } 

Professor Streitberg follows his theory 
through the other cases—locative, vocative, 
accusative—with great success. The short 
form of the suffix in the vocative arises by 
its throwing forward the accent : zarnp but 
matep. But whether this change of accent 
position is due to enclisis of the vocative 
in principal sentences, as Streitberg following 
Hirt is inclined to believe, is another story. 
One most ingenious point is made in the 
treatment of the accusative. The question 
is raised: why Bov=Skt. gam, Zjv=Skt. 
dya@m with circumflex? The answer is: 
From an original *géuom, diéwom would come 
*goum, *diéum as monosyllables ; the mora 
represented by -y- is lost, and according to .« 
the second part of the rule the circumflex 
appears instead of the acute accent. On the 
other hand there is no lengthening in 
accusatives like dda, rouséva or Hyenova, for 
in these no mora is lost ; the final syllable 
is still represented by -a. A long form of 
the accusative is found only in such stems 
as carry a long vowel throughout the whole 
paradigm ; aifwva, genitive aifwvos, ke. 

One curious result may be mentioned in 
passing. Streitberg declares himself a con- | 
vert to the theory, to which Brugmann has 
clung throughout amid almost universal 
opposition, that original 6 in an open 
syllable is represented in Skt. by a. 

The treatment of the genitive is inter- 
esting. It has often been held on the 
ground of the requirements of the Teutonic 
and Slavonic languages that there must 
have been a genitive suffix in -so. The 
present theory requires that the -so-suffix 
should be original. The treatment however 
of the genitive *podo-so seems to me incon- 
clusive. Streitberg contends that this form 
had the principal accent on the first syllable, 
the last syllable disappeared having no 
accent, while the middle syllable having 
some amount of accent survived; hence 
*podos out of *pdddso. This contention 
seems at variance with the rest of the 
theory ; at any rate it requires, in order to 
be convincing, a fuller treatment than it 
receives, 

The verb is a less fertile field than the 
noun, but here too the theory obtains some 





THE CLASSICAL REVIEW. 


interesting results. ‘The 3rd person of the 
perfect yéyove is the same as the Skt. 
jajana ; the Skt. Ist person jajdéna should 
be represented in Greek by *yéyeva. The 
Skt. causative bhdrayati represents by its a 
the same original sound as in gop- of dopéw ; 
the Aryan passive aorist which appears only 
in the 3rd person singular (Skt. wvdci, Zend 
avaéet &e.) is explained, after Osthoff, as a 
substantive form of the same type as 
otpodi-, tpoxi- in Greek, which, like other 
noun forms, has engrafted itself upon the 
verb. The only part of the verb where the 
working of the theory is certainly manifested 
is in the -s- aorists. The suffix is shown to 
occur in three forms: -es- -as- and -s-; but 


117 


the vowel is recognized as in reality part 
the root; Skt. andisam is ' 
veloped from *a-nayi-gam. Other points r 

garding the verb cannot be illustrated fron 
the classical languages and may be left | 

the reader to find for himself. — 

In conclusion it may be said that Professor 
Streitberg’s style is a model of lucidity, that 
the paper contains many points which have 
not been mentioned here, and that the conclu 
sions, if they can on the whole be held to be 
established, as I think they can, make it the 
most important article that has appeared 
on such subjects for a considerable time. 


P. GILEs. 


therefore at 





MERRY’S EDITION OF THE WASPS. 


THE latest instalment of Dr. Merry’s 
edition of Aristophanes maintains the high 
level of its predecessors. It displays the 
same literary skill in translation, the same 
refined scholarship, and the same delicate 
appreciation of wit and humour. 

To the man of the world, who wishes to 
renew his acquaintance with the greatest 
works of Aristophanes, no edition can be 
more highly recommended. As_ school- 
books, Dr. Merry’s volumes have the merit, 
which is rarely met with in present-day 
text-books, of not superseding the use of 
grammar or dictionary. They are suggestive 
without being exhaustive. 

The Vespae has of late attracted much 
attention from scholars. The edition of 
Mr. B. B. Rogers is familiar to all lovers of 
literature. During the last few years have 
appeared the commentaries of Dr. Blaydes 
and Prof. Van Leeuwen, and recently the 
school-editions of Dr. Merry and Mr. 
Graves. After the exhaustive work of Dr. 
Blaydes, editions for some years to come 
will be but tend from his detrva. Nothing 
is left to commentators but the labour of 
judicious selection. His edition is a mine 
of learning. To one who wishes to tread 
with a firm foot the mazes of Attic diction, 
his work is indispensable, although the 
redundancy of expression and the extraor- 
dinary haste with which he has thrown his 
materials together sometimes so exasperate 
the reader that he is tempted to cry out 
(with a slight variation of Socrates’ words) 
GdAws pe mpodidacke iva pi) ard cod dmogot- 


tyow. Indeed, in the absence of an index, 
his commentary is :— 


Monstrum horrendum informe ingens cui 
lumen ademptum. 


Still, although one of the extremities of 
this monster may be Poditwov, assuredly 
the other is not of brass, but of gold. 

The edition of Mr. Graves can be com- 
pared more fairly with Dr. Merry’s. It is 
equally scholarly, though perhaps it falls 
below the other in literary taste and 
originality. However it possesses the merit 
of giving the chief MSS. variants at the 
foot of the page, and of marking the 
Strophes and Antistrophes in the choral 
odes, and the divisions of the Parabasis. 
The omission of all such marks is the great 
defect of Dr. Merry’s edition. It is a work 
of some difficulty to discover where the 
Strophes end and the Antistrophes begin. 

To the schoolboy no doubt, for whom 
this edition is primarily intended, the 
metrical character of a choral ode is one of 
the mysteries of Faith. Credunt quia non 
intellegunt. In Dr. Merry’s edition the 
lines are cut into apparently equal lengths, 
but, on a closer examination, this superficial 
correspondence is found, in many cases, to 
fall to the ground. A few instances will 
prove my point. 

Vesp. 296— 


7 7 > _ - 
’A| orpaydaAous Syrovbev, © Tat. 


118 


in the Strophe is supposed to correspond 
with line 308— 


, 7 / 
zopov EXXas tepov. 


Quite apart from the question of metre, 
Hermann’s eipety improves the sense. No 
doubt, this is an instance of the error called 
haplography. 

Vesp. 407 foll.— 


A 5 A A > An 
VUV EKELVO VUV EKELVO 
> iA 
TovévOupov, @ KoraLo- 
fd 2 i) 
peaOa, Kévtpov évrérat O&v. 


The last line must correspond with line 


465— 
ws AdOpa y’ éAdpBav’ brodca pe. 


This passage requires very careful treat- 
ment. As the other lines, in this part of 
the chorus, are trochaic, there must be 
something wrong with both 409 and 465. 

I believe that Blaydes is right in rejecting 
6€¥, Which is obviously due to dfv4upov, and 
in reading éxreracOw. 

With regard to line 465, which is much 
too long and redundant in expression, I 
believe that Ad@pa is a gloss on td in 
trvotea, and that édévOavev of BCSV is right. 
So I suggest :— 


¢ A 
ws éhavOavey trioted p: 


or, if we keep dév, ws eAdvOavev y’ k.7.A. 
Vesp. 410— 


kal keAever’ adtov HKELW 
e a= 3, , 

as éx’ avopa picdmoAw 
dvta KaTroNOvpeVoV, OTL 
tovoe Adyov cia héEpet, 

c ‘ ‘ ld tA 4 

os xpy by Sikalew dikas. 


must correspond with Vesp. 468— 


» >» , 
OUTE TL EXwV TpOpacw 
» , > ‘ 
ovte Adyov eitpameXov 
aiTos dpYwv povos. 


The first thing to be said with regard to 
these passages is that the last three lines of 
the Antistrophe are excellent Cretics, and 
that the last three lines of the Strophe must 
be made to correspond. This can be 
done without much difficulty. és yp7 must 
be rejected, with many editors, as a gloss. 
In line 412, I suggest kdroXovpevov o71— 
"Ovra is not required. For the short 
syllable at the end of a Cretic, compare 


THE CLASSICAL REVIEW. 


Ach. 301 
para. 

Two lines must have been lost in the 
Antistrophe corresponding with 410—11. 

In line 411, I read pucddnpov with Herm. 
to make the verse trochatc. 

Vesp. 418— 


lal lal lal 7 
KATATEAW TOLOLV imTEVTL KQTTU- 


® TOAL Kal Oewpov PeowwexOpia. 
This is a very strange Cretic. Read 
moXus. 


Vesp. 474~-- 
gol Aoyous, & pucddnpe Kal povapxias epacta. 


This is supposed to correspond with line 
417 :-— 


A aA > 3 \ ‘\ , 5 2) , 
Tavta, Ont ov dewa Kal Tuparvis oT eudhavys ; 


Dindorf’s épdv is a very easy correction. 


Vesp. 385— 
AéE | ov pos evvovs yup dpacets. 
must correspond with line 343 :— 
ore A€yes Tu TEpl Tav ve | av. 


Xv, or a similar word, must be inserted 
after Néyers. 

It is very difficult to understand what Dr. 
Merry’s views are with regard to Strophic 
correspondence. In these days of metrical 
heresies, one could sympathize with an 
editor who gave up the whole question in 
despair. But Dr. Merry is more sanguine. 
On line 339 he says ‘There must be some- 
thing wrong in kai tiva rpdpacw exwv as the 
line does not correspond metrically with 
GAN éraye tiv yvaBov’ (370). The corre- 
spondence seems to me perfect. In his text 
he prints the line without xa/, thus de- 
stroying the correspondence. Again, on 
line 526, viv 5) rov ex Ojperépov, he says 
‘ydv dy is introduced for metrical reasons, 
viv dé R. and V.’ Verily, this is straining 
at a gnat. In the very next line he prints 
without remark the unmetrical yupvaciov 
Neyewv Te Set to correspond with 632 ovdevds 
nKovoapev ov | de. Obviously we must read 
yupvactov det Tu N€eyew. 

Again, 638 Koved mapndOev, dor eywy’ 


must correspond with tévde A€yew: spas 
yop ws (532), Read xotr. Again, 749 


meopevos Té cor is supposed to corre- 
spond with 736 od d€ wapov déyov. The 
emendation is obvious, viz. 7iOdpuevos. Indeed, 
apart from metrical reasons, this is the more 














- non-fulfilled condition. 
passage in Xenophon to bolster up this 


THE CLASSICAL REVIEW. Lit 


common form in Aristophanes. 886 must 
be read as an Iambic line (with Porson), 
eivexa, being corrected to évexa. The same 
inconsistency is to be observed throughout. 
At times, Dr. Merry’s chivalrous trust in 
the impeccability of the copyists is pathetic, 
though misplaced. He seems to think 
with Vivien ‘ Unfaith in aught is want of 
faith in all.’ In this spirit, he prints 
nxnkoew and jv (although R. gives 7, Vesp. 
1091) as lst persons singular. Has the 
teaching of Cobet and Rutherford been writ 
in water? In Aves 511 rovri rotvw oix 
nocw ’yo is the reading of most MSS. 

27 Sewov yé rov’or k.7.A. Surely rover is 
right. As we were taught at school, yé rou 
assents with a limitation, and may be 
translated ‘at any rate.’ The stock quota- 
tion used to be Vesp. 934 érywe ye tou. In 
many places Dr. Merry maltreats these 
particles, e.g. 912 read euot yé row (not euou- 
yé to). In lines 25 and 155, he gives the 
unmetrical iddvte Ttovodttov évirvoy and 
dvAatré 6 drws, without comment. I have 
no doubt that A. Palmer’s rotro toivirvov 
is right in the former passage, and that, in 
the latter, we should read (with Elmsley) 
poxArod | didar6’. 

Vesp. 471, he reads :— 


ay > »” , \ a os a 
éo@ orws GveEv paxys Kal THs KaTogElas Bo‘s 
és Aoyous EAPoipev GAAHAOLTL Kat Siadrdayas ; 


In his notes he says dézws is not a final 
conjunction, but equivalent to gua ratione, 
and for the opt. he refers to the well-known 
passages A. Agam. 620, and E. Ale. 52. 
Now, it must be noted that this construc- 
tion only occurs in tragedy, and, with the 
single exception of Alc. 52, in negative 
sentences. If ever there was a certain 
emendation, it is Hermann’s dv éx payys. 
Its truth is shown by line 866 or yevvaiws 
€k TOD ToA¢uov | Kal Tod vecKous EvveByrTov. 


Vesp. 602— 
oxepat 8 amd tov dyabdv oiwv doKdeles Kal 
- KATEpUKELs. 


This is Dr. Merry’s text, which is very 
hard to translate. Bergk’s emendation 
oxevat 5€ ?-dowv &. olwv 7’ K.7.A.—80 far, at 
least, as dowv is concerned,—seems to me 
certain, 

Vesp. 709, he refuses to accept dv0 pupiid’ 
av tov Sypotiav ewv év racr aywous—the 
easy correction of Dobree—on the ground 
that dy is not necessary in the apodosis of a 
He appeals to one 


heresy. Cobet has taught scholars how 
very unsafe a guide Xenophon is in 
question of Attic usage. It is, moreover, 
by no means certain that dv has not dropped 
out after noxvvopnv in the passage quoted, 
Vesp. 694, Kad’ os mpiové’ is more than 
doubtful, as Aristophanes invariably uses 
OoTEp in a simile. The sole exception 
quoted is 1490 TTT EL Ppvviyos ws Tis 
aXextwp, Which, as Blaydes suggests, may be 
a quotation from Phrynichus. In his 
note on 795 Dr. Merry ingeniously defends 
i) & os A€yov, quoting épy A€yow from 
Herodotus. But [ know of no instance of 
A€ywv with 7 8 ds, and, on general grounds, 


yeAov makes much better sense. At other 
times, Dr. Merry is not true to his 
allegiance. Vesp. 121 the vulgate dre diy 88 


(ci. Heel. 195, 315, 827 et passim) for 6re 
djta must be recalled: 606 xdred’ jxové’ 
must be restored (from R.) for kar’ eloyjxov6’ 
(a rare word not used in prose). 


ae , , “ ; : -o 
651, vOooovV apXatav €v TY) ToAE €VTETUAULOP. 


Dr. Merry accepts Reisk’s em. for MSS. 
evreroxuiav. The .shortening of the &@ is 
supported by analogy, but I do not see how 
the vulgate is impossible, if we translate : 
‘the disease that has littered in the state.’ 

1157. Dr. Merry accepts trodvov for 
azodvov, but in 1168 inconsistently keeps 
trodvodpevos, although he reads birodyjcacbat, 
against the MSS., in 1159. 

1309. He rightly reads veorAovTw Ppvye 
for tpvyt. Sometimes Dr, Merry, like the 
Atxaos Aoyos, throwing off his ipdrioy, 
éEavropoAe to the opposition. On 13540 he 
accepts an emendation of that arch radical, 
Dr. Blaydes, viz.: ot« draow dots éorty, 
(MSS. otk dweor; rod ’oriv). He would 
have been wise to have followed that 
scholar in some other passages. Dr. Blaydes 
has succeeded by a change of a single letter, 
or by a slight alteration of the punctuation, 
in greatly improving the text of many 
passages. To take the changes of punctua 
tion first. Vesp. 385. Read muwvvos xal, 
pavOdver; av te mdOw “yd. 448 remove ; 
after Onpiov : otherwise ofd' dvapnvnabes has 
no construction. The note of interrogation 
must be placed after «lvac in 451. 773. It 
simplifies the construction to place the pion 
orrypy after Kxabrjpevos. Otherwise the sen: 
tence is too loaded (dav 0€ vidy...vorTos)- 
Line 937 the troorrypy must be removed 
after mapeivat, aS pdpTupas is the predicate. 
Otherwise robs pdprupas would be nec 8sary, 
957. Read 6 te; cov mpopdxerat. 1184 the 
brooriypy must be removed after xorpoAcyy, 


120 


and placed after éby. "Edy tT kotpoddyw is 
bad Greek. Again, in many cases the 
change of a single letter improves the text. 
Vesp. 61 évaceAyawvopevos (MSS. dvacedy. cf. 
eumapowelv). 110. w’ éxou dixalov (MSS. 
duxalew). 125. éEadpieuev (Nauck for MSS. 
‘ecadpeiopev). 128 éereBicapev (MSS. eveBv- 
wapev). 201. riyv doxov (MSS. rH doxd). 
‘B19. Om’ dkovwv or éeraxovwv (MSS. izaxovor, 
which is unmeaning here). 353. éaia (MSS. 
driav). 422. xaloé y’ adtots éfododpev (MSS. 
attis, Which is not a Comic form). 452. 
GAN adés pe kat od Kal ov (MSS. aves). 
544. Padrrodopor xadovpeGa (MSS. xadoipefa). 
676. upxas oivov (so the line is cited by Pollux; 
MSS iUpyas, otvov), 790. xdaer’ évéOyKe 
(MSS. éréOnxe—cf. évOeors ‘a mouthful’). 
978. airetobe (MSS. aireire, which cannot 
mean ‘supplicate’). 1032. of devorepar pev 
am’ éh0artpov Kuvyns axtives €Aaprov. A 
brachylogy of comparison (MSS. deworara). 
1107. dorep és tavOpyvia—(MSS. dozepei. 
"AvOpyviov is the nest not the hornet). 
1116. wdvov Katecbiovew (MSS. yévov). 
1418. pi pn Kadrtéon (MSS. xadréons. zpoo- 
kadovpor occurs in the preceding Ime), 
1436. otros (MSS. ards). 1118. tebewpyxa 
ovdapot (MSS. otdapod). 

With reference to Dr. Blaydes’ emenda- 
tion of Vesp. 21 mpoBadre tis rota. cvpmdrais 
Aéyov—(MSS. zpocepet) I feel inclined to 
say, as Bentley said of one of his own 
emendations, aut ita scripsit Aristophanes 
aut deliravit. Blaydes’ note shows that 
mpoPdadAev is the word invariably used of 
propounding a ypidos. If it is right, how 
can we explain the blunder in the MSS.? 
Dr. Merry’s rpoepeé will not do, especially 
with Aéywy at the end of the line. Ipoa- 
yopevw means ‘to proclaim.’ I had thought 
of mpodépe, which sometimes means to 
propose, as the future does not seem to be 
necessary since Aristophanes is quoting a 
riddle which was commonly proposed at 
dinner-parties. But zpoBadre is much 
better. So much for the text. 

I have noticed very few errors or omissions 
in the notes. 

Vesp. 3. The imperfect of rpoode/Aw should 
be zpoaderes, not zpovderes. I do not see 
what there is to object to in the present mpot- 
peters. It is not every dpa that is followed 
by an imperfect. 10. rrépevoy, not rrdpevor, 
is the Comic form. 12. vverakrjs is an 
imitation of a Persian proper name. I 
have heard it translated ‘from the land of 
Nod.’ 40. Bédeos also=stupid (so Schol.). 
58. There is no note on the Schema 
Pindaricum. 92. éxet could not mean ‘ during 
his nap’—Rogers is right. 151, There is 


THE CLASSICAL REVIEW. 


no note on Kazviov. As the poet does not 
seem zapatpaywoetv, 1 suggest tod Karviov, 
201. zpooxvAtcov is an impossible accentua- 
tion, 221. 7d woré with the future is very 
strange. It ought to mean jam dudum. 
We may compare Lucan 2, 524 yam dudum 
moriture. 242, It ought to be noted that 
this is the only passage (Zr. 48 is doubtful), 
except in the phrase xOés te kali zpwyy, 
where x6és occurs in Aristophanes in 
iambies. I believe that KAéwy is a gloss, 
and that we should read éy@és péev oty 6 
kndenav. (I now see that this is Meineke’s 
reading.) 250. The Scholiast recognizes 
only rpdpvgov. 243, zovypdy does not mean 
‘merciless’ but ‘not of the best quality,’ as 
neEpav Gpynv tptov Shows. Aristophanes can 
never resist a reference to the xkpoppvoé- 
epvypta. 304. The proverb cixa airety is not 
explained—oixa payeiv = tpudav (so Schol.). 
334, tadra eipywy could not mean ‘bar from 
this,’ but ‘in this way.’ 3857. toyvov atrds 
€u“avtov, as the parallel passages show, does 
not mean ‘ was master of my own actions,’ 
but ‘was in the plenitude of my powers.’ 
460. dp’ éeuehropev rofl ipas amocoBynce ; 
Dr. Merry propounds the strange heresy 
that dpa is not an interrogative, and 
compares Ach. éue\ov apa. The position 
makes all the difference. 661. It should be 
noted that xarafés always means ‘to pay 
money’ not ‘to set down’ and that éviavrod 
has the article, which is indispensable, in 
all MSS. I believe that Blaydes’ emenda- 
tion, which occurred to me independently, is 
right—dmé tovtov viv toto. dixagtais Ges 
pucbov tod y eviavtod. Kara may be the 
the remains of -xacrais. 691. ‘dpaypyyv has 
uniformly & in Aristophanes.’ This sentence 
must be withdrawn. Blaydes gives some 
instances: Pax 1201; Plut. 1019. There 
is no objection to the lengthening in 
Anapaestic verse. 771. ‘There is no ety- 
mological justification for the jingle between 
yAvdoe. and 7Auov.’ Dr. Merry forgets that 
nAidlecGar also means apricari. 917. No 
notice is taken of the Scholiast’s explanation 
of kowds=Kowwvds. Cf. Oed. R. 240. 955. 
olds te moAAois tpoBatias edheotava. Dr. 
Merry has a strange note ‘usually some: 
part of the verb eiyvé is introduced to- 
complete this phrase, but not always.’ In 
that case there would be no connecting 
particle. The line means ‘and qualified to” 
ete. 


Vesp. 1208—9— 


BAE, zat’ GdAXd devpi KaraxAwels tpopavOave 
Evprrotikes €ivar Kat EvvoutiagTiKos. 





THE CLASSICAL REVIEW, 


Tt has not been remarked by any com- 


‘mentator, so far as I know, that Bdelycleon 


is here affecting the style and language of 
the Athenian jewnesse dorée, which Aristoph. 
Knights satirizes, 1378 foll. :— 


A / 
AH. ouvepxtixos yap €or Kai mepavricds 
‘\ ‘ 5 
Kal YVWLOTUTLKOS Kal oadijs Kal 


KPOVOTLKOS 
KaTahyTTiKos 7 apicta tod GopvBy- 
TLKOV. 
ATOP. ovxovv katadaxtvAtKds av Tod AaXdn- 
TLKOD. 


1397. There is no note on the very extra- 
ordinary form @vyarepos. If it is right, we 
must hold that the dptowwdis speaks, in 


Epic phrase, like the d \ughter of a queen 
1434. § Well you must bear in mind the 


« “wWwral , ° 
answer he gave.’ dA\’ ofy means ‘at any 


sors ; ° 

rate.” As airds cannot mean ‘he, we must 
accept Blaydes’ ofros. 1454. It must be 
noticed that all MSS. have }. Dr. Merry’ 


account of the readings of the MSS. is not 
quite correct. 1455, For éyo cf. Soph. dd 
@V €XOL TE KAL OUVALTO, 
in maxims. 1483. ‘The mischief is indeed 
spreading’: xai 57) means ‘ already.’ 

Sed haec quidem hactenus. Dr. Merry’s 
notes are so admirable in style and matter 
that it is very difficult even for the most 
captious critic to find anything that provokes 
comment. 


It is generally used 


W. J. M. Srarkie. 


GRAVES’ EDITION OF THE WASPS. 


The Wasps of Aristophanes, edited by C. E. 
Graves, M.A. Cambridge University 
Press. 3s. 6d. 


Wiru this addition to previously published 
literature on the subject, English students 
may consider themselves exceptionally well 
off in aids for studying this amusing and 
instructive play. The purpose of the present 
article is not so much acriticism on a volume 
which it is doing bare justice to the editor 
to praise for thoroughness and accuracy, as 
a discussion of a few places on which it 
appears to me that light may yet be cast. 
Any criticism I might offer would be of the 
nature of a wish that Mr. Graves would 
have more often the courage of his opinions, 
and instead of withholding his vote between 
two rival views (both admirably stated), or 
merely expressing a doubting preference, 
would boldly plunge in his Yjdos on one side 
or the other xara yvopny tHv dpiornv. The 
answer, no doubt, may be made that in a 
book intended for students what is required 
is that the evidence should be placed before 
them, and that premature and one-sided 
views should be discouraged. But experience 
seems to show that students are too apt to 
take the view that points about which their 
teachers tell them that no one knows for 
certain, are points about which they need 
not concern themselves. Not only is the 
definite view more interesting, because it is 
more definite, but because polemics have an 
interest of their own. Apart from the 
‘exigencies of the student, it seems to me 
‘certain that progress in knowledge is best 


- 


secured, not by neutrality, but by careful 
and conscientious advocacy. We get an 
approach to a solution of a difficulty when 
we find an increasing preponderance of 
opinion, independently formed, in favour of 
one possible reading or explanation over 
others. 

T take as a first illustration v. 36, €xovra 
huvyv éumempnpevys bos, on Which Mr. Graves 
says: ‘The meaning [of ¢urerpyuévys| ap- 
pears to be “blown out, bloated’’ (schol. 
éumepvonperyns, Taxeias): but it might be 
“ burnt, scalded.”’ I think the second alter- 
native is made impossible by the tense ; for 
though a pig might make a considerable 
outcry during the process of singing, it 
would probably become quieter when the 
process was accomplished. Buttmann, 
Lexilogus p. 484 (E.T.), quotes a gram- 
marian’s note in Hesychius, d@ev xai jects 
Tempnevous Tors Tedvonpevors. Perhaps 
however Meineke is right in keeping the 
accusative éumempynévyy with R and V, 
which with @wviv must mean ‘ at full blast, 
with no possible ambiguity. 

Take again vv. 1020 sqq. 

od’ ef tus dpacrys 
xwpwderrOar madix’ <avtod pucay tomevde mpos 

avTov K.T.A, 

Does xwpwdeicGar depend on picoy or ob 
Yorevde? Mr. Graves says that ‘ either view 
can be supported grammatically,’ but allows 
that the latter agrees best with line 1028 


‘va tas Movoas alow xpiytat my Fpoaywyors 
, 
aropyvy: 


122 


Surely the case can be put more strongly. 
The question is practically: does the poet 
claim credit for refusing to insert, or for 
refusing to suppress, something? He says 
that if be had either (a) suppressed, or (b) 
inserted, the matter in question, his Muse 
might be open to the charge of being zpoa- 
ywyos. It is impossible to me to see how 
his Muse could in any case become zpoaywyos 
by omitting something. Therefore the poet 
claims credit for refusing to insert some- 
thing. Therefore kwywdeicba depends on 
eovrevoe Yather than on pucav. 

In his text Mr. Graves is very conserva- 
tive; some will perhaps think unduly so. 
He retains line 135 éywy tporous dpvaypocep- 
vakous Twas in its place, notwithstanding the 
difficulties (which he acknowledges) both as 
to grammar and sense presented by the line. 
Whence the line came it is impossible to 
say; in any case we should, I think, read 
dpvaypoceuvixovs, the word being an example 
of the forms in -xés that were part of the 
slang of the time (cf. Hg. 1378 sqq.). 

At line 713 Mr. Graves prints the un- 
metrical oipor tito’ borep k.7.X. of the MSS. 
in spite of strong claims (fortified by Suidas) 
of the correction oipo. ti rérové’ ; darep 
KetaN. 

Difficulties in the text reading are always 
pointed out in the notes, and proposed cor- 
rections quoted. One or two readings quoted 
might either have been passed over in silence, 
or quoted only to be condemned. Atv. 61 
van Leeuwen’s 


ovd’ abfis ave yavodper eis Hipuridny, 
-and at v. 155 the reading 
pvrarré & orws py Tv Badavov éexrpwkerat 


both ought to be put out of court by the 
break between the two short syllables of an 
anapaestic foot. 

The scene vv. 136-229, as arranged by 
Mr. Graves, requires four speaking actors. 
It seems probable, however, that Bdelycleon, 
when he utters v. 138 

ov mrepiopapetrar opav TaXéws Seip’ &repos ; 
is still off the stage. Sosias then runs 
‘detpo, that is, off the stage, and to 
Xanthias at v. 142 is assigned the charge 
of watching the door. After a short pause 
the actor who has been playing Sosias pops 
his head out of the chimney as Philocleon. 
Meanwhile Xanthias on the stage continues 
to watch the door, and it is he, not Sosias, 
who says at v. 152 dd riv Ovpav db. To 
which Bdelycleon replies wile viv ofddpa, 


THE CLASSICAL REVIEW. 


and promises to come to his assistance. 
Lines 202-210 should be assigned to 
Xanthias and Philocleon, as they are by 
Dindorf. In this way the whole of the 
scene can be managed by three speaking 
actors, and there is no place in the play 
where more than three are required. I 
think Sosias nowhere appears after v. 138. 
Mr. Graves thinks that he appears in the 
trial scene (vv. 891 sqq.) as a kwgpov tpocwz7or, 
but there is no evidence of this. 
On v. 151 


A \ A: , / 
OOTLS TATPOS VUV Kazviov KekAnoopat 


Mr. Graves observes ‘The joke on xaavy or 
kamvos is obvious, but its further point is not 
so clear. A rough-tasting wine, or the vine 
producing it, is said to have been called 
kazvias, but this line very likely originated 
the idea. It seems plain that calling a man 
Kavos Was a gibe of the day, see 325, and 
this may be all that is meant. The scholiast 
says that Ecphantides, an old comic poet, 
was nicknamed Kazvias, ‘old smoky,” when 
dull and stupid. Rogers supposes that some 
disreputable Athenian named Capnias may 
be intended.’ 

It seems to me that xazvias is a natural 
enough word to describe wine of a particular 
flavour, and that it is quite unnecessary to 
suppose that this line originated the title, 
and rather hard to see how it could. Kazvias 
oivos is mentioned by Aristophanes’ senior, 
Pherecrates, and that the epithet was already 
used of a kind of vine is clear from vy. 
325-6, where we are told that Aeschines, 
who is kamvos, is (no true kamvios dpmedos 
but) a wevdapdpags. Nor can I believe 
with Mr. Rogers that Kazvias was the real 
name of an Athenian. The gibe imputes a 
combination of real meanness with pretended 
grandeur. It is given to a man who brags, 
(‘ gasses’ some would say now), about his 
castles in Spain—or in Thessaly (v. 1246). 
Compare the Latin fumum vendere. The 
notion of dulness or murkiness in a poet 
has nothing cognate with the situation ; so, 
though the scholiast may be right about 
Kephantides, it is unlikely that heis referred 
to here. Two persons nicknamed xazvos or 
karvias at this time are mentioned in this 
play ; Proxenides at v. 325 and Aeschines 
vv. 325, 459, 1220, and 1246. The great 
probability is that it is one of these two men 
who is here (v. 151) referred to. Aeschines 
appears to be a demagogue, from the com- 
pany in which we find him in v. 1220. He 
is the son of ‘Sellus,’ that is of a mendicant 
impostor. At line 459 we have Sellus 


i ae 





THE CLASSICAL REVIEW. 123 


comically expanded into Sellartius, where 
the second element probably is dpros, bread ; 
though it is just possible that it may contain 
a reference to the real name of Aeschines’ 
father, whatever it was. Was it worth 
while to commemorate, as Mr. Graves does 
on v. 459, the scholiast’s notion that 
Sellartius is connected with céAas ! 

At v. 1267 we have a mention of Amynias, 
who, as Mr. Graves says in his note on v. 
1265, had wasted his substance by luxury 
and gambling. His note on v. 1267, 

GAN ’Apvvias 6 SéAXov paddov obk Tov 

Kpwfvrov, 
runs thus: ‘ Amynias was really the son of 
Pronapes (75) and is here called son of 
Sellus, to show that he was a brother xarvés 
of Aeschines.’ But the cases of Aeschines 


and Amynias were totally different 
were ‘ brothers’ in no sense. Aeschines was 
born poor and bragged of wealth he never 
had ; Amynias was born rich but spent his 
money, and there is no evidence that he eve 
bragged of possessing wealth after he had 
lost it. I believe that pa@dAov is here used 
(as paAdov dé is more usually) as a corrective 
The poet having in mind Amynias’ destitute 
condition, by a pretended slip of memory, 
ascribes to him a beggarly origin, a mistake 
which he immediately corrects. Under 
standing oxatds édofe we get the meaning: 
‘but (I can’t say so much for) Amynias, the 
beggar-born—no, I beg his pardon, one of 
the race of the ancient Top-knot.’ 
E. 8. THompsoy. 


Lhey 


STRACHAN-DAVIDSON’S CICERO. 


Cicero and the Fall of the Roman Republic, 
by J. L. Srracwan-Davipson, M.A., 
Fellow of Balliol College. Oxford. 
‘Heroes of the Nations’ series, G. P. 
Putnam’s Sons, New York and London. 
1894, 5s. 


It is with sound judgment that Mr. 
Strachan-Davidson has described Cicero as a 
hero of the Roman Republic. Cicero believed 
that a ‘Free State was the only form of 
government worth having’ and died a 
martyr to that confession. This must lead 
us to form a judgment of Cicero very 
different from that which the ponderous 
learning of Drumann and the passionate 


genius of Mommsen have made fashionable. 


Mr, Strachan-Davidson says it is impossible 
to escape from the influence of the latter 
great scholar: it is certainly difficult, such 


are his learning and lordly power, but Mr. 


Strachan-Davidson has escaped it; and 
with deep sympathy for one who reverenced 
his conscience as his king, and with a 
strong love of freedom, he has written in an 
exceptionally charming style an Lhrenrettung 
of Cicero the politician which serves to place 
him in his proper position as the faithful 
servant and defender of the Roman Republic 
in the last days of her existence. In a truer 
sense than either Brutus or Cassius was 


Cicero the last of the Romans. 


To describe all the excellences of the 


- work would be an endless undertaking ; but 
some especially suggestive discussions may 


be mentioned. Mr. Strachan-Davidson 
points out the absence of organized ‘party’ 
at Rome in Cicero’s time except among the 
revolutionary faction, and shows that it was 
asindividuals rather than as party supporters 
that men were advanced to office (92). 
Nothing could be better than the contrast 
between the projects of the impatient, 
shortsighted Catiline, * who did not draw the 
sword before he blew the horn,’ and the 
patient, far-seeing Caesar (117-9). Very 
true also is the psychological contrast 
between  self-conceit—obstinate, perverse 
and a sign of weakness—and the exuberant 
and innocuous peacock-like vanity of mere 
loquaciousness, which may exist with true 
greatness of character (193). It might 
have been added that the latter quality, 
though not the former, is generally found 
with almost complete absence of jealousy ; 
and we think that Mr. Strachan-Davidson 
has failed to emphasize with sufficient force 
this beautiful trait in Cicero’s character. 
Very different were the majority of the 
Roman nobles and their natural leader, 
Pompey. Again, Cicero's personal loyalty 
to his friends, e.g. Sestius, Flaccus, should 
have received more recognition, especially 
as Pompey’s faults in that respect meet 
with well-deserved castigation (265). It is 
pointed out with all possible detail (109, 
260, 274) that it was the constant effort of 
Cicero to effect the natural union between 
Pompey and the body of nobles ; that the 
latter, like true oligarchs, were suspicious 


124 


and jealous of one another; and that yet 
Cicero, well knowing all Pompey’s faults 
and meannesses, was yet instinctively drawn 
to him even to the last by an irresistible 
attraction—‘ein dimonischer Zug,’ as 
Weidner puts it. 

With genuine poetical feeling does Mr. 
Strachan-Davidson point out the tragedy of 
Cicero’s political life: the union of Pompey 
and the nobles was at last effected when it 
was too late and Caesar was already march- 
ing on Rome (321) ; ‘ the union of the orders’ 
and the ‘consent of Italy’ were realized 
at the very end when the Republic had only 
raw recruits as defenders and well-trained 
veterans as opponents (406). And it is 
with true feeling for Cicero’s character, 
who even more than Caesar had much in 
his soul besides the statesman, that Mr. 
Straechan-Davidson adduces as an effective 
reason for Cicero’s joining the triumvirs in 
55 the fear that, if he did not join them, he 
would compromise the fortunes of his 
brother Quintus (208); and that he asks us 
to reflect how proud a moment it must have 
been for Cicero when he could say officially 
to the senate ‘the legion which was com- 
manded by Lucius Piso, one of Antony’s 
lieutenants, has gone over to my son Cicero 
and placed itself at his disposal’ (390). 
The true reason why Pompey joined Caesar 
in 60 is- doubtless that given by Mr. 
Strachan-Davidson (206) viz. that he did 
not think Caesar was a military genius or 
would ever dream of carrying out his 
political designs by an armed force; why, 
not even in 51 did the nobles or Cicero 
think it. The discussion on the religious 
veto and its origin is excellent, not merely 
in statement of fact but also by reason of 
the delightfully cynical account of the way 
in which the Romans ‘manufactured good 
luck’ (211). Highly ingenious too is the 
explanation given of the fact that the 
violence of vituperation is much less in 
modern times than in ancient Rome, viz. 
that a powerful restraint has been put upon 
the spoken word by the practice of duelling 
(257). Mr. Strachan-Davidson thinks (411) 
that the Liberators were short-sighted in not 
waiting for Caesar’s death in the course of 
nature, and perhaps then the army, like that 
of Cromwell, would have passed over to the 
legitimate government. They doubtless 
would have replied to such a suggestion 
that if they waited Caesar would have time 
to train a successor. However it is for the 
thorough insight into Cicero’s character 
that we must be most grateful to Mr. 
Strachan-Davidson. He holds that Cicero 


THE CLASSICAL REVIEW. 


had a sound conscience which was perplexed 
by too subtle an intellect. When he 
perceived his duty plainly he boldly and 
unreservedly faced it ; but he generally saw 
so many sides of a question that he was 
unable to decide quickly as to what course 
to pursue (426). It might have been added 
that it was mainly this characteristic,— 
which was a defect in Cicero regarded as a 
practical politician,—that combined with his 
great gift of language to make him pre- 
eminent as an advocate, and as a popularizer 
among the Latin nations of the many-sided 
philosophy and culture of the Greek world. 

The character of Pompey too, who 
‘considered himself a privileged person’ (42), 
is sketched with great insight. He cer- 
tainly had not ‘ the vulgar ambition to make 
himself a despot’ (87); and we do not 
believe that in the proposal of Messius 
(248) Pompey had any further motive than 
the desire to be as fully equipped as possible 
in order to carry out the work he had to do 
speedily and effectively. There was nothing 
Pompey liked so much as doing an easy 
thing with great éclat. He quite wished to 
be loyal to the state, he did not dream of 
military tyranny, and we think that the 
nobles knew it; Cicero certainly knew it 
(Rull. ii. 62). This gives a satisfactory 
reason why they always treated Pompey 
with such mean jealousy and frustrated his 
legitimate demands for grants of land for 
his soldiers. But we fail to see the drift of 
Mr. Strachan-Davidson’s persistent attacks 
(167, 347—351) on Caesar’s military des- 
potism as an evil. Of course it was an evil, 
but a lesser evil than the continuance of the 
senatorial government. Against whom are 
these attacks directed? If they are meant 
for Caesar’s modern flamen, Mommsen, 
they but reflect what he has said with all 
the emphasis in his power in one of the 
most celebrated passages of his History (iv. 
466-7 E.T.). And there is something 
further. The Empire had to be organized. 
It is only by despotism, under whatever 
form it may be veiled, that deliberate 
organization can be effectively carried out 
on a large scale. As well blame our 
government of India as blame Caesar’s 
government of the Empire or any other 
order of discipline to which human beings 
are compelled to submit in order that their 
development may be more steady or their 
process of decay less painful. Nor can we 
altogether agree with the views maintained 
of Caesar’s proposal in the matter of the 
Catilinarian conspirators (141 ff.). Excite- 
ment ran high; but Caesar kept his head 





THE CLASSICAL REVIEW, 


and made a proposal that was undoubtedly 
within the law, which was perhaps unduly 
tender to the Roman citizen, but still the 
law. The apparent severity of that proposal 
is to be explained by the fact that nothing 
lenient would have been listened to for a 
moment: but of course Caesar knew that, 
once the excitement was over and the 
conspiracy finally crushed, the conspirators 
would at least be given a fair trial; for he 
knew that the final overthrow of the con- 
spiracy was but a question of time now 
that the conspirators at Rome were arrested, 
whether they were executed or not. 

We regret that we do not hear more 
about Cicero’s private life: also that so 
little is said about Cicero’s encouragement 
of, and kindly feeling towards, young men, 
and his relations with the brilliant Curio. 
Marcus Brutus too receives scant recogni- 
tion. Mr. Strachan-Davidson might have 
taken heart of grace and acknowledged the 
genuineness of most of the Epistles to 
Brutus, now that even Meyer has struck 
his colours. Some mistakes of a very 
trifling nature have been made, e.g. the 
senator with whom Catiline placed himself 
was M. Metellus, not M. Marcellus (123). 
The account followed in the narrative of 
the deaths of Quintus and his son (80) is 
that of Appian (iv. 20) not that of Dio 
Cassius. Negotiator is rather. ‘money- 


125 


lender’ than ‘trader.’ The letter to Trebi 
anus (Fam. vi. 10) mentioned on p. 355 
belongs to the autumn of 46, not to the 
beginning of 45. There is a slight incon- 
sistency in what is stated as to Cicero's 
desire for the augurate on p. 218 and p. 
290. On p. 49 for ‘path’ read ‘way of 
glory.’ 

The translations (which are rightly 
numerous) are of such a perfect nature that 
it would require a Cicero to praise them, e.g. 
Jautriz suorum regio ‘this clannish land of 
ours’ (8); aqua haeret ‘there is a stoppage 
in the current of my action’ (266) ; desiderio 
pristinae dignitatis ‘from a sense of the 
aching void left by the loss of my old 
independence’ (358). 
ableness’ is a_ better 
than ‘fluidity’ (197); valde bella (195) is 
rather ‘quite charming’ than ‘mighty 
fine’; and we altogether object to ‘ piracy’ 
as a rendering of Jatrocinium (128, 258): 
‘pirate’ is confined properly to a robber on 
the high seas. 

The work is enriched with many pictures 
of places and people (mostly from Duruy) 
and of coins (from Cohen and Babelon). 
There is a map to illustrate the War of 
Mutina, and a copious index. In truth a 
book ‘ omnibus numeris absolutus.’ 

L. C, Purser. 


Perhaps ‘ impression- 
word for mollitiam 


FREESE’S TRANSLATION OF ISOCRATES. 


The Orations of Isocrates, translated by 
J. H. Freese, M.A., formerly Fellow of 
St. John’s College, Cambridge. With 
Introduction and Notes. Vol. I. London : 
G. Bell and Sons, 1894. 5s. 


Isocrares, it is believed, has never yet been 
translated into English; and Mr. Freese 
may be congratulated on the clear field thus 
opened to his labours. Like the best of the 
Bohn series, and notably C. R. Kennedy in 
his Demosthenes, he has produced something 
much better than the ordinary ‘crib’; a 
version at once literal enough to afford a 
model for students, and readable enough to 
attract the English reader, now supposed to 
be more intent than formerly upon the 
masterpieces of ancient literature. The 
present volume contains the first ten orations, 
including the important political pamphlets 
Panegyricus, Philippus, Archidamus, Areo- 


pagiticus and On the Peace, but not extending 
to the Antidosis or the still more interesting 
six forensic speeches. Nearly half the 
volume (as far as § 70 of the Phi/ippus) had 
been translated by Mr. A. H. Dennis of 
Exeter College, Oxford, who was prevented by 
professional engagements from carrying out 
his plan. His MS. was handed over to Mr. 
Freese, who has revised and completed it, 
and has added the Introduction and Notes. 
These, it must be admitted, are somewhat 
thin in character, but criticism is to some 
extent disarmed by the candid confession 
that they are intended mainly for English 
readers. In the second volume we are 
promised some Appendices dealing with the 
Athenian constitution and other matters ; 
when it is to be hoped that the writer will 
aim at a more solid contribution to scholar- 
ship. We have read the notes throughout, 
and have compared Mr. Freese’s English 


126 


with the Greek in the three speeches 
Panegyricus, Philippus, and Areopagiticus. 
The general result, as has been said, is 
nighly satisfactory : we have noted some few 
points as needing correction. 

Paneg. § 126 ody ws éxeivoy rA€omev warep 
mpos deorroTyV GAAnAwY Katyyopycovres; Mr, 
F. gives a figurative sense to rAéopev, and 
translates: ‘Are we not drifting into his 
hands as into those of a master, ready to 
blame each other for the result?’ The 
meaning surely is: ‘Do we not sail to him 
(i.e. send embassies) as to a master, in order 
to accuse one another (as before a judge) ?’ 
Isocrates is describing the abject attitude 
of the Greeks in general towards the Persian 
king after the Peace of Antalcidas. § 151 
éferalouevol mpdos avtots tots PBacrdciors, 
‘searched on entering the palace,’ is ex- 
panded into ‘ they are subjected to inspection 
on the very threshold of the royal palace.’ 
Besides its diffuseness, this translation em- 
phasizes the wrong part of the sentence. 
§ 153 érépov tocovrov xpdvov is not ‘for 
twice that length of time’ (which would be 
sixteen months) but ‘for as long a period’ 
(viz. eight months). § 180 xa@ OAs ris 
‘EAAddos ‘concerns the whole of Hellas.’ 
Rather ‘is directed against the whole of 
Hellas.” § 187 riv & eddaipoviav tiv ex ris 
Actas cis tHv Etpwrny diaxopioamer: eddat- 
povia is here ‘ wealth,’ not ‘happiness’ as 
Mr. F. translates. Phil. § 5 eis otdviep 
Aakedaipdviot Kupnvaious amroxicay ‘such as 
the place to which the Lacedaemonians have 
removed the Cyreneans.’ The reference is 
not, as the note states, to the planting of 


THE CLASSICAL REVIEW. 


colonies from Cyrene, but to the colonization 
of Cyrene itself by the Lacedaemonians. 

80 xowds is translated ‘accessible to all.’ 
It may be rendered in one word, ‘ impartial.’ 
§ 130 For ‘whom I hope will be best able 
to do so’ read ‘who I hope’ &e. § 152 
‘Having practised upon them, gained ex- 
perience, and come to know what manner of 
man you are.’ This would be a rendering 
of yvods otos ef, if self-knowledge were in- 
tended : the Greek is yvwo6eis ofos «7, and 
the meaning ‘recognized for what you are.’ 

Areop. § 38 the notes require revision : 
the age of the Ephebi and the explanation 
of émi dieres HBjoa are not given according 
to the latest authorities ; and the Areopagus 
was not called 7 avw BovdAr ‘as holding its 
meetings on the hill,’ but as enjoying an 
honorary precedence like ‘the Upper House.’ 
§ 46 Kat ratra vopobernoavtes ovde Tov 
Aourdv xpdvov @Avyopovv: ‘And while they 
made these regulations they did not neglect 
the future’ is not a happy rendering, sug- 
gesting as it does that tov Aouroy xpovov is 
accusative after dAvywpovv. Of course airav 
must be supplied: ‘they did not neglect 
them later on.’ § 58 ocuvédpovs kai ovy- 
ypadéas ‘committees and boards,’ The 
former word may pass muster; the ovy- 
ypadets Were commissioners with full powers 
(airoxparopes) to draw upa new constitution, 
Thucyd. viii. 67. 

We had marked the following passages 
as examples of Mr. F.’s best style: Paneg. 
§ 185, Areop. §§ 37, 53. But we must 
abstain from giving extracts. 

W. WaytE, 


HOESS ON THE STYLE OF ISOCRATES. 


De ubertate et abundantia sermonis Isocratet 
observationum capita  selecta  scripsit 
GuILeLmus Hogss. Friburgi Brisigaviae 
ex officina C. A. Wagneri mpcccxcit. 
Pp. 56. 


Tuis is a statistical investigation, undertaken 
at the suggestion of Otto Hense, of the 
tautological use of synonyms by Isocrates of 
which Dionysius of Halicarnassus speaks de 
Isocr. ¢. 3, ba07 avaykn taparAnpwpac. Aé~ewv 
ovdev Hpedovody xpjobar Kai dropnktvew Tépa 
TOU xpyoipmov Tov Adyov, and again 7d KdAXos 
THS amayyeNlas év TO TepitTa TUepevov. That 
Isocrates chooses his words with care and 
shows discrimination in the use of words 


almost but not quite identical in meaning 
has been pointed out and illustrated by Dr. 
Blass (Até. Bered. iit 125 sq.), who agrees 
with Kyprianos in seeing here the influence 
of Prodicus. But it is also true that 
Isocrates (who is imitated in this by 
Demosthenes and by Latin writers) fre- 
quently, for the sake of rhetorical effect, 
uses two or more expressions where one 
would have sufficed. Thus, as Hoess points 
out (p. 13 sqg.), sometimes the same thought 
is expressed both affirmatively and negatively, 
as in xvi. 44 008’ dvayxacbeis add’ Exdv, 
sometimes two or more synonyms are joined 
by kat (or ov—oidé, py...undé) or less often 
aovvderws, aS in Xil. 264 éryvovv, elypdovr, 





THE CLASSICAL REVIEW. 


épaxapifov, sometimes words scarcely differing 
at all in meaning are joined by # or by xat 
...KQl, OUTE...OUTE, pHTE... pyre, aS in Vv. 136 Kal 
Tiv oTpareiav roveiobar Tavtyv Kal ToAELELV Kal 
Kuvouvevey, Sometimes again a word like xaAds 
or péyas is followed by one of narrower 
meaning, as in vi. 32 papripia peclm Kal 
cadeotepa. 

Hoess has made a complete collection of 
all such instances as occur in the Isocratean 
‘corpus,’ giving separate lists of substan- 
tives, adjectives, adverbs, and verbs for each 
oration and epistle, and adding a table 
showing the total number of instances. 
There is of course room for much ditference 
of opinion as to whether a particular word 
or phrase should be regarded as redundant, 
but Hoess has chosen his instances with care, 
availing himself of the investigations of 
H. Schmidt in his Synonymik der griechischen 
Sprache. The collection will therefore be 
of value to all students of Greek synonyms. 

But Hoess has made a further use of his 
results. He regards the relative frequency 
of synonyms as a norm of style which may 
serve as a subsidiary aid in fixing the date 
of publication of certain orations and in 
deciding as to the genuineness of doubtful 
works. He therefore fixes the dates of the 
orations as exactly as possible, following in 
the main Dr. Blass’s Afttische Beredsamkeit 
ii.2, and then shows how far his statistics 
serve to confirm these dates. The table on 
p- 43 shows that Isocrates made but slight 
use of these synonyms in his earlier works, 
that he used them with greater frequency in 
his middle and best period, and that they 
are found most abundantly in his latest 
works. 

An example or two will make Hoess’s 
method of proceeding clear. The date of 


127 


the Helena (Or. x.) is, as is well known. a 
matter of dispute. While Blass (/.¢, p. 122) 
regards it as one of the earliest works of 
Tsocrates, Professor Jebb, whose careful 
discussion of this point is not referred to by 
Hoess, thinks it may probably be put al out 
370 B.c? (Attie Orators iio pp 102 sg.), and 
Keil (Anal. Isoc. p. 8} dates it circ. 365 B.c. 
Now the synonym test points to a date not 
only later than the Busiris (before 380) but 
also later than the Plataicus (374-372), 
Hoess therefore dates it (p. 47) ‘ certe non 
ante a. 370,’ though he had previously (p. 
6) on other grounds said that it was written 
‘paulo ante ol. 100(380).’ Again, as to the 
date of the Archidamus (Or. vi.) opinions 
differ widely, Blass (/.c. p. 263) placing it 
after 356, while Professor Jebb refers it to 
366, and Keil to 365 B.c. Hoess prefers the 
earlier date and seesa confirmation of it in 
the small number of synonyms as compared 
with the latest orations. 

The question of the genuineness of the 
Ad Demonicum and of the Letters is dis- 
cussed by Hoess at some length. He is 
convinced that the former is not the work 
of Isocrates, though he seems to have 
overlooked the De Tsocratis _Demonicea 
of Ponickau (noticed in the Classical Review 
iv. p. 422). On the other hand he considers 
that the question of the genuineness of the 
Letters needs a far more thorough investiga- 
tion than it has yet received. The ninth 
letter he declares to be undoubtedly spurious 
(p. 7), but does not venture to give a decided 
opinion as to the others. And indeed, as 
all the letters are accepted by Blass as 
genuine and rejected by Keil as spurious, 
we may well agree with Hoess that ‘adhue 
sub iudice lis est.’ 

Henry CLARKE. 





HULTSCH ON THE TENSES OF POLYBIUS. 


Die erzihlenden Zeitformen bei Polybios, ein 
Beitrag zur Syntax der gemeingriechischen 
Sprache, von Frieprich Hu.tscu. Leip- 

zig, 1891—1893. §8. Hirzel. 11 Mk. 


THE learned editor of Polybius has set 
himself the task of illustrating that writer’s 
use of the historic tenses, and with 
characteristic industry has collected some 
six thousand instances which heendeavours to 
reduce to their proper categories. He forcibly 
points out that the interest and value of the 


language of Polybius consist in the fact 
that he heads the list of the writers of the 
common dialect (ow). The later writers 
of Greek in the imperial age attempted to 
recall the grace and elegance of the Attic of 
the best period, but only sacrificed vigour 
and life. Polybius, the practical statesman, 
sought the ground-work of his style in the 
living speech of his time, and was acceptable 
to his contemporaries in proportion as he 
was intelligible. It is easier, as Hultsch re- 
marks, to note the differences in the language 


128 


of Polybius from the Attic than to form a 
comprehensive view of the principle under- 
lying it ; and it is this latter that he has 
attempted to do. He therefore goes to the 
root of the matter, adopting the logical 
division of G. Curtius between the time- 
period (Zeitstufe) and the time-mode 
(Zeitari) of a tense. By the former an 
action is indicated as past present or future, 
by the latter as accomplished, continuing, 
or beginning. And he lays it down as a 
preliminary formula that by the imperfect 
indicative an action is conceived as con- 
tinuous, by the aorist as transient (dauerlos). 
But the use of the one or the other (though 
not uninfluenced by objective rules) depends 
mainly on the subjective judgment of the 
writer. Consequently verbs which in 
themselves involve an idea of duration will 
mostly be found in the imperfect,—as for 
instance, in military matters dye, tpodyew 
and other verbs of going, starting, and 
sending are often found in the imperfect 
where the aorist might have been expected ; 
those that involve the idea of suddenness or 
brief duration will mostly be found in the 
aorist. After devoting three sections to a 
dissertation on the three kinds of imper- 
fects—of duration, of development, of 
description—and their interconnexion, he 
treats the case of verbs of attempting and 
undertaking, such as d.ddvar in the sense of 
‘to offer,’ we(Gev of unsuccessful advice, 
repacbat, éyxepety and the like; and next 
shows by a comparative table of imperfects 
and aorists how the former predominates in 
connexion with adverbs or adverbial 
expressions. Similar lists are given of verbs 
of demanding and exhorting, as, d&vody, 
rapawveiv, tapakadev (p. 109, 110); and of 
sending or dispatching (p. 120—122) ; and 
of numerous others. Though the instances 
are all but exclusively taken from Polybius, 
these two hundred large pages ($$ i.—xxii.) 
contain really a most valuable treatise on 
the use of the aorist and imperfect applic- 
able to Greek generally. The second part 
($$ xxiii—xxx. pp. 179) treats in a 
similar manner the interchange of the 
imperfect and aorist, starting with an 
exhaustive examination of the usages of the 
verbs, yiverOa, exe, Netrew, peverv, Hevyew 
and BddXev, with their compounds. In 
§ 28 the use of the aorist to express 
continuance for a definite space followed by 
change of the action conceived, as in 
éréuewe Tpets Hucpas (21, 43, 9), roddv pev 
xpovov yropyoav (1, 10, 3), is illustrated at 
some length, as well as the prevailing use 
of the aorist with certain adverbs, as Tédos. 


THE CLASSICAL REVIEW. 


Finally, Hultsch gives a list of verbs whose 
aorist and imperfect tenses have the same 
form in the third person singular, as 
éyeipen, KAtvew, Kpivew, KTeive, mpavyely, 
diapOecpew and others, the time-value of 
which therefore has to be settled by the 
context. In § 29 he further illustrates the 
variation of the use of the imperfect and 
aorist by examining in detail certain 
descriptive passages, such as for instance 
the account of the battle of Cannae (3, 1138 
—116), which he speaks of as ein stilistisches 
Meisterwerk, and points out the delicate 
shades of meaning expressed by the respect- 
ive uses of aorists and imperfects in the 
words: éxiveu—zapevéBade—eSeratre— Karte 
otnoe. He gives moreover a long list of 
phrases in which the aorist and imperfect 
occur in clauses joined by «at or by the 
particles pev and d¢, when a similar differ- 
ence in the meaning is required. This is 
farther illustrated by contrasted clauses in 
§ xxx. ; and he then goes on in §§ xxxi. and 
XXxii. to discuss and illustrate the uses of 
the historic present and pluperfect. The 
historic present is naturally of much rarer 
occurrence in descriptive passages than the 
imperfect and aorist, the proportion in the 
first book being about one to sixty, and 
somewhat lower still in the following four 
books, and in the fragments less frequently 
by about a half than in the first five books. 
As to this Hultsch remarks that as 
Polybius occupied many years in the com- 
position of his work, and consulted a large 
variety of authorities, some variation of 
style between the beginning and end of his 
history was to be expected. But neither 
in his use of the historic present nor in that 
of the pluperfect does he differ materially 
from the usage of his predecessors ; and the 
most valuable part of the last section is the 
list of passages in which clauses are found 
joined by xai, pev...d€, re...re and other con- 
junctive or disjunctive words, containing 
respectively a pluperfect and imperfect. As 
the three parts of the treatise were printed 
originally in the transactions of the philo- 
logisch-historischen Classe der Kénigl. Sachs- 
ischen Gesellschaft der Wissenschaften, the 
pagination is not consecutive, but a fairly 
good index at the end of the third part 
helps to obviate any difficulty which might 
be caused by this. There can be no doubt 
that by his laborious, and it is to be feared 
unremunerative, work the author has 
earned the gratitude of students of Polybius 
and of Greek generally. 

| E, 8. SHuckBureH. 





































oe ounty : 


THE CLASSICAL REVIEW 129 


HARDY’S CHRISTIANITY AND THE ROMAN GOVERNMENT. 


Christianity and the Roman Government, by 
E. G. Harpy, M.A. London. 1489. 5s. 


Mr. Harpy’s interesting little volume is, like 
many other recent bits of writing, an outcome 
of Prof. Ramsay’s stimulating book upon 
The Church in the Roman Empire. But it 
is to Mommsen’s article ‘ Der Religionsfrevel 
nach romischem Recht’ that Mr. Hardy is 
chiefly indebted for his point of view, which 
may be thus summed up in his own words 
(p- 101): ‘Christianity by virtue of its 
inherent disobedience (obstinatio, rapdtaéis) 
was a criminal offence, but in the eyes of 
the police administration, not of the law.’ 
And on p. 90 he argues that ‘when 
repressive measures were taken, they would 
be taken usually, not from any “ Flavian 
policy” (i.e. general instructions given to 
provincial governors to put down Christ- 
ianity), not because membership in the sect 
was looked upon as treasonable by the 
government, certainly not because the Church 
was looked upon as “an organized unity 
dangerous to the state,’ but in consequence 
of some manifestation of hostile feeling on 
the part of the populace.’ 

This criticism of Prof. Ramsay (whose 
are the words italicized) is probably true of 
the period ending about 250 a.p., but 
surely not after that. However this is a 
minor point, and his readers have reason to 
thank Mr. Hardy for the clear and cogent 
way in which he brings out many points, 
e.g. that the Church was not persecuted in 
Bithynia or elsewhere because it was an 
illicit collegium ; that torture was, primarily 
at least and in theory, applied to Christians 
in order to induce them to recant and save 
themselves from the death-penalty ; that 
the obligation to sacrifice to the Emperor's 
image was a mere test to determine whether 
men were truly Christians ; that the Roman 
authorities, during the first two centuries 


at any rate, rather reined in than not the very 
intelligible fury and rage which the new 
religion excited in the breasts of men and 
women in general. Had the populace been 
allowed to have their way with the religion 
from the first, it would | 
survived, 

Mr. Hardy thinks that the whole of the 
charge against the Christians was from the 
first persecution under Nero onwards simply 
and solely that of hatred of the human 
race, ‘odium generis humani.’ ‘They were 
potentially outlaws and brigands and could 
be treated by the police administration as 
such, whether in Rome or the provinces.’ 
He cannot therefore agree with Prof. 
Ramsay that the Flavian emperors intro- 
duced any new principle of punishing the 
mere profession of the name of Christian. 
That name simply denoted the bearer’s 
mischievous tendencies. 

In an appendix Mr, Hardy reprints the 
Acts of the martyrs of Scili and the Acts of 
Apollonius. He accounts for the action 
taken by the Senate in the latter trial by 
supposing that the martyr was a senator. 
This Mommsen in a recent paper read 
before the Prussian Academy denies. The 
circumstance, which I noticed in repub- 
lishing these Acts, that the Armenian MSS. 
name Tarruntenus and not Perennis as the 
prefect in charge of the case, has not been 
considered either by Mommsen or Harnack 
in their arguments against and for the 
senatorial rank of the accused. Yet it 
must surely bear upon it, besides putting 
the date of the martyrdom two or three 
years earlier. 

Mr. Hardy’s book may be warmly re- 
commended to all who desire a guide to the 
intricate question of the origin and nature 
of the earlier persecutions, It is a pity 
that no index of contents is supplied. 

Frep. C, ConyBEARE. 


hardly have 





TWO BOOKS ON HORACE. 


Scholia Antigua in Q. Horatium Flaccum, 
Vol. I. Porphyrionis Commentum recen- 
suit A. Hotper. Ad Aeni Pontem, sumpt- 
ibus et typis Wagneri. MDCCCLXXXXIIII. 


suchungen von G. FriepRicH, Gymnasial- 
oberlehrer zu Schweidnitz. Leipzig: B. 
G. Teubner, 1894. 6 Mk. 

NO. LXXVI. VOL. IX. 


Tue first of these two books is a most 
careful and admirably printed edition of 
Porphyrion’s commentaries according to the 
text of the Codex Vrsinianus (Vaticanus 3314), 
which the editor carefully collated during 
October and November 1887, and which he 
assigns to the ninth century. All varia- 
tions from the actual text of the MS. are 
accurately marked, any letters which are 
K 


130 


added, removed, or corrected being clearly 
indicated, and various readings and emen- 
dations being given at the foot of the page, 
while an elaborate index of more than 
100 pp. makes the book most useful for 
reference. As it stands the work is in- 
dispensable to all advanced scholars, but 
even for them it would be improved—and 
it is the only improvement needed—if the 
editor had prefaced his work with some 
observations on the value which those who 
consult them may reasonably attach to 
Porphyrion’s comments and on the authority 
which his evidence as to a reading 
possesses. 

A well-known instance will show that 
some guidance is needed. In Od. 1, 20, 10 
his reading is tw wiuis wuam, but Sat. 
2, 2, 48, commenting on the words [nfamis. 
Quid? Tum, he writes ‘ Figura nota apud 
H. transeundi ad aliam rem... ut: 


Caecubum et prelo domita(m) Caleno 
Tu(m) bibes wua(m),’ 


where although, as in domita and wua, the 
final m is not written, yet the context shows 
that ftwm must be read, while in the note 
on Od. 1, 20, 10 there is no indication 
whether twm or tw is intended. Again in 
the difficult passage Sat. 1, 4, 14 (Crispinus 
minimo me provocat) one would be glad to 
know what is the weight of his testimony 
to Latin usage when he says that mimo 
me digito provocat is proverbial ‘ cum volumus 
quem intelligt tantum valere minimo digito, 
quantum aliwm totis viribus,’ because, if his 
opinion is worth anything, this is a case 
where we must accept it, or else with 
Bentley write nummo ,. for the alternative 
explanation ‘lays me long odds,’ though 
generally received, is impossible, the person 
who ‘lays long odds’ surely ‘challenging 
with a great sum’ and not ‘with a very 
little one.’ But, though in the first of these 
passages Porphyrion has preserved the only 
possible reading, and also by the spelling 
of his MS. illustrated the origin of the 
hopeless tu bibes, and in the second passage 
provides an explanation which is at least 
probable, still those who accept him as an 
authority will find themselves often very 
wrong. What can be said for a man who, 
in Ep. 1, 11, 26 non maris effusi late locus 
arbiter, explains arbiter as = remotus—‘ maris 
arbiter est locus, qui trans mare longe positus 
atque discretus est’—and then adds potest 
arbiter et medius in’ellegi ? Or who remarks 
on tratis natus paries... that poets were 
‘accustomed to cover the walls with wax so 


THE CLASSICAL REVIEW. 


that they might write down on them any 
idea that occurred to them in the night’? 
Finally, what real knowledge can a man have 
who, Hpod. 17, 52 utcunque fortis exilis puer- 
pera, takes exilis not as exsilis but as exilis ? 
Perhaps, however, this last instance might 
do some good if it could help to show the 
need of some sane method of Latin spelling. 
Doubtless in a book of this sort the MS. 
spelling should be preserved. It is odd 
enough—first wiuis then bibes, furnos and 
Jornos in the same line, yperbolen, lagynae, 
lymfa, cumuae, celeps, Saffo, Cantauer, &e.— 
but it has a historic and technical interest, 
while its eccentricities can do no harm. 
Yet, if a scholar like Porphyrion—or Por- 
fyrion—can be seduced by a merely am- 
biguous spelling into such a blunder, it is 
natural to ask whether the absolute chaos 
which now prevails as regards Latin ortho- 
graphy is not a grave danger to classical 
study. By all means let experts, who 
choose, study the subject. Their labours 
may ultimately produce some good, and a 
careful examination of the new Corpus 
proves that they have at any rate established 
one law, to wit ‘that in the Augustan age 
no two poets spelled alike and no individual 
poet had any fixed standard of spelling of 
his own.’ This happy result may amuse 
cynics or delight the learned, but those who 
are concerned with classical study as a 
means of education know that the absence 
of any common system of spelling in ordinary 
books is a constant hindrance to their work 
and productive of no good whatever. The 
difficulty did not exist thirty years ago when 
Latin spelling was possibly bad, but at any 
rate fairly consistent. Now, under an 
affectation of knowledge, every editor spells 
as he chooses or as accident dictates, and 
one more stumbling-block is wilfully placed 
in the way of classical teaching in all our 
schools.! 

The second book is eminently interesting. 
It is a remarkable combination of sound 
scholarship, acute criticism, and foolish 
speculation. The writer is one of those 
scholars who find more in the Odes than is 
reasonably to be got out of them, but who 
in the course of their investigation bring 
out very forcibly many interesting details. 
Like most critics of this class he takes a 
special interest in the unknown ladies to 
whom Horace addresses Odes, and on his 


1 Take these instances from the Corgus Poetarum : 
Aen. 8, 173 agnoscere, 180 adgnovit, 6, 580 anticwm, 
648 antiquum ; Hor. Od. 4, 8, 24 opstarect, 9, 43 
obstantis. Any decent ‘reader’ would correct such 
blunders during printing, if only he were allowed. 


age rsa Sete t=. 


iis > 


































THE CLASSICAL REVIEW, 131 


first page he gives us a very vigorous sketch 
of Lyde, whose identity in the three Odes 
addressed to her is so clear that, he says 
‘few things can be affirmed with greater 
certainty,’ though how his statement that, 
after being ‘ wooed’ as a ‘coy young maid’ 
by Horace in 3, 11, she becomes in 3, 28 
‘seine Geliebte’ can agree with Nauck’s 
famous description of her there as ‘ eine 
fleissige ernstgesinnte haushilterische Schaff- 
nerin, is a question which no one less versed 


in such matters than German professors can 


be expected to decide. On 2, 12, 28, how- 
ever, he shows sound judgment in preferring 
the MSS. occupet to Bentley’s occupat, which 
Schiitz defends on the ground that to 
describe Licymnia as ‘letting the same 
kisses, which she facili saevitia negat, be 
snatched from her and sometimes snatching 
them herself’ is ‘nonsense.’ ‘ Yes,’ is the 
reply, ‘nonsense undoubtedly, but the 
charming nonsense of passion. With rapere 
occupat the passage becomes—as happens 
with most conjectures in Horace*—more 
intelligible and more home-baked (haus- 
backener).’ So, too, he will have nothing 
to do with the absurd theories that magis 
poscente (poscente abl. abs. Kiessling ; = a 
poscente Schiitz) do not go together, because 
to say ‘more than you who beg for them’ 
would represent Maecenas as a very cold 
wooer. He shows by some most interest- 
ing quotations how passionately devoted 
Maecenas was to Terentia, and, after re- 


viewing her relations to Augustus, rightly 
says: it must have 





‘As matters stood 
sounded like music to Maecenas’ to hear 
her passion for him described as greater 
than his own. Similarly in 1, 6, 28 he 
rightly rejects both the usual explanations 
of sectis wnguibus as either specially ‘ pared’ 
so as to do no harm or specially ‘sharpened’ 
so as to do a deal. The adjective only 
describes the ordinary carefully trimmed 
condition of a fashionable beauty’s nails 
(the opposite of Canidia’s irresectwm pollicem), 
though of course it suggests that the wounds 
they will inflict in their ‘fierce’ attack will 
be only trivial. Surely this is sounder 
sense than Bentley’s appalling stric/is which 


he ealls certissima emendatio. 


In 2, 11, 4 nec trepides in usum is ex- 
plained ‘and be not slow to enjoy.’ Of 
course trepidare is often almost timere, and 
soin Lue. 5, 728 dubiwm trepidumque in 
praelia, where the context makes the sense 
certain, the word describes’ not ‘quivering 
eagerness ’ but ‘ trembling hesitation.’ Here, 
however, the context absolutely forbids our 
taking the word in such a non-Horatian 


sense, for treprde 8 in usum is carefuily 
opposed to remittas quaerere (‘slack thy search 
and be not eager’). Moreover, why should 
any one show ‘trembling hesitation’ ab ut 
enjoying life? ‘Trembling hesitation’ alo 

going into battle is natural, but with regard 
to doing something pleasant the phrase is 
absurd, In the same stanza, however, the 
difficult Scythes : . Hadria divisus obyje 
is well explained by the remark that the 
Adriatic is considered as a shield which j 
held close to the body (objecto nobis, not 
Scythis), though the enemy, as here the 
Scythians, may be at a considerable distance, 

Epist. 1, 20, 19 cum tibi sol teypidus plures 
admoverit aures. The school in which the 
book is being read is out of doors, ‘a hedge- 
row school,’ cf. Dio Chrysost. 20, 9, of ray 
ypapparwv didacKxador pera tov traldww ev rads 
odots Kd@yvra, and as the day grows warmer 
idle loungers come and listen to the school- 
master, and it is to these gentlemen he is 
to recite the passage which tells them how 
the writer is, like themselves, a humbly 
born, little, irritable, placable, sun-loving 
fellow. j 

Od. 2, 18, 29 Orci fine, about which 
editors make so much ado, is clearly shown 
to be merely the Greek Gavarou réAos repro- 
duced on the principle stated A. ?. 52 et nova 
Jictaque nuper habebunt verba fidem si Graeco 
fonte cadent parce detorta. In 1, 20 he 
shows that Peerlkamp’s correction of sud- 
movere litora to promovere litora would pro- 
duce a meaning the exact opposite of what 
Peerlkamp intended, cf. Sen. nat. quaest. 
3, 27, 10 of an inundation jam enim pro 
movet (mare) litus, 28, 3, litus prolatum ; 
but the process of killing conjectures and 
emendations is like the labour of Hercules 
against the Hydra. 

On Od. 1, 15 there is a long discussion to 
show that this ode is closely connected with 
the obviously political ode which precedes. 
Thus the shepherd Paris naturally becomes 
M. Antony, and Helen is Cleopatra (as 
Cruquius suggested), Menelaus disappears 
because the name was commonly used for an 
injured husband (ef. Cie. ad Att. 1, 18, 3), 
and so could not be applied to Octavian, 
who therefore is represented by Diomede, 
Ulysses is Agrippa, the ‘wrathful fleet of 
Achilles’ is the tleet of 8. Pompeius, and so 
on, and so on, It is as clear as daylight. 
The last line ignis Jliacas domos with its 
well-known metrical difficulty hides a mystic 
meaning. After nineteen centuries the seer 
has seen the interpretation thereof. Jgnis 
with a final short! Yes, O blind leaders of 


the blind, but at last there has yo u 
K 


clo 


132 


Gymmasialoberlehrer, who is a teacher in- 
deed. Put an x before J/iacas, and what 
then ? Niliacas domos—‘ Rome shall burn the 
palaces of the Nile.’ No prophecy was ever 
plainer. 

In Lpod. 9 Biicheler’s view that Horace 
was present at Actium is rightly accepted, 
the words fluentem nauseam thus obtaining 
some real force which in old editions they 
never had, and the hopeless sinistrorswm 
becoming explicable as a pictorial word used 
by one who had actually just seen the vessel 
moving off ‘ to the left.’. On 1.17 he remarks 
that ‘all MSS.’ have j/rementes (and not 
the ace. form frementis) which must there- 
fore go with Gali. One would like to know 
the value of this assertion about frementes 
being necessarily nominative. Common sense 
almost demands its connection with equos. 

Od. 3,19. The scene opens with Horace 
strolling with some friends who are discuss- 
ing antiquities, when Horace, who is cold 
and hungry, suggests that they should dine: 
as the writer well puts it, ‘nowadays he 
would have suggested an adjournment to 
some well-known restaurant,’ but in those 
days a detrvov amo EvpPoddv was the only 
method. In 1. 9 he makes Horace staré with 
three bumpers of unmixed wine, for he says 
the ‘mixing’ first begins inl. 11. Thus he 
makes Horace begin with pure wine, and 
then recommend wine mixed with either 75 
or 25 per cent. of water. Did ever mortal 
man drink liquor on such principles? It 
would be folly for an English editor to refer 
a German critic to an English edition, for 
German students of Horace respect nothing 
but conjectural emendations, and scorn every 
one but Bentley, but I may perhaps refer to 
Marquardt ed. 2, p. 335, where the view, 
which I arrrived at independently some 
time before, is taken—viz. that the ‘rule of 
3 or 9’ does not refer to the proportions of 
wine and water, but to the number of cyathi 
taken in each bumper to drink to the various 
toasts. 

Od, 3, 11, 49. He rightly calls attention to 
the pathos of dum favet nox et Venus, used 
just when the lovers are parting, and to the 
wonderful music of the words 7 secundo. . . 
with the almost rhyming assonance of se- 
cundo and sepulchro. Has not the remark- 
able hiatus in secundo omine, which no editor 
notices, also something to do with the charm 
of these lines ? 

Od. 4, 11, 2. Fulges is probably /futwre— 
‘there is parsley for weaving garlands, and 
ivy with which you will be able to deck 
your hair,’ ef. fervit in the oldest MS8S., Od. 
4, 2, 7. In Od. 3, 8, 5, utriusque linguae 


THE CLASSICAL REVIEW. 


refers to Latin and Etruscan, for ‘in what 
G'reek folk-lore would Maecenas find anything 
about a bachelor keeping the purely Roman 
Matronalia?’ On the other hand, Etruscan 
‘tradition’ (sermones) was the basis of many 
Roman ceremonies. Od. 2, 13, 7 he proves 
to demonstration that robur is the Tudlianum 
to which the conquered chief was led away 
as the triwnphator began‘ to ascend the 
Capitol, on the summit of which he halted 
HExpis Gv TOD oTpaTHyov TOV ToAEUiwy Oavatov 
amayyetAn mis (Jos. de Bell. J. 5, 6). Od. 
3, 1010 ne currente retro funis eat rota is 
of a man pulling up a weight with a pulley, 
and, on finding it too hard a task, letting 
the rope go, but the application is not, as 
most take it, to Lyce, but to Horace, who, 
as the last lines show, threatens to give up 
the weary task of wooing her. 

The twelfth Ode is described as an elabo- 
rate ode to be sung on the marriage of 
Marcellus and Julia. Starting on this as- 
sumption, the writer finds all sorts of hidden 
meanings in the various names mentioned, 
e.g. Fabricius and Curius suggest Pyrrhus, 
Pyrrhus suggests Epirus and the Hast, and 
so the reference to Octavian crushing the 
forces of the East on the coast of Epirus is 
at once obvious. The curious list of names 
in the ninth stanza—Romulus, the ‘ quiet 
reign’ of Numa, ‘the haughty fasces’ of 
Tarquin and Cato’s ‘famous death ’—sug- 
gests the ideas of ‘ fratricide, ignominious 
repose, tyranny, and civil war,’ and so 
affords a contrast to the glorious era which 
is just commencing. Of course it is impos- 
sible to reply to such theories. There is 
undoubtedly in some of the odes a meaning 
not apparent to the ordinary eye, and 
scholars from time to time flatter themselves 
that they have discovered its exact character. 
But after the lapse of 2,000 years, the ele- 
ments of uncertainty in such investigations 
are so great that exactitude is impossible, 
and those who attempt to fill in the histori- 
cal background to an ode with too careful 
detail are in danger rather of marring the 
effect of the ode itself than of bringing it 
out in clearer relief. When we conceive the 
ninth Epode as written at sea immediately 
after Actium, we at once double our power 
of understanding it, and similarly in Od. 1, 
12 the knowledge that it was written for 
the marriage of Julia and Marcellus would 
place us in a better position for understand- 
ing it, but, even then, it would be folly to 
inquire exactly what suggestion—if any— 
Horace intended to convey to Augustus and 
his court by the mention of ‘Regulus and 
the Scauri.’ 





THE CLASSICAL REVIEW, 135 


Od. 1, 37, 14, lymphatam is not vano 
pavore territam, as in Orelli, but is exactly 
parallel to dulei fortuna ebria and quidlibet 
umpotens sperare expressing her ‘ delirious 
hopes,’ and so in exact opposition to veros 
timores. Od. 4, 8, 24, the name Romuli is 
put designedly at the end of the sentence, 
because Horace wishes to suggest that 
but for the help of potentes vates the 
very name would have been forgotten. 
The famous difficulty in 4, 8, 17 is really 
a joke! The confusion between the two 
Scipios is designedly introduced to show 
the importance of not leaving your fame and 
name to be the sport of vague tradition. In 
]. 25 of the same ode, virtus is comnected 
with vatum, a discovery which is not new 
among us ‘remote Britons,’ though German 
editors ‘almost unanimously’ connect the 
word with Aeacum. Od. 3, 18, the festival 
of Faunus referred to is not that in Decem- 
ber, but in February, because the kid born 
in February could not be ‘of a full year’ 
before then! One would like to know why, 
in face of the fact that this ode is obviously 
written for the December festival, cf. 1. 10 
cum tibi nonae redeunt Decembres, we are to 
be compelled to accept the dictum of any 
editor (e.g. Kiessling) that pleno haedus anno 
go together, or indeed are Latin. But it is 
foolish to ask questions. Horace knew less 
about his own works than his critics do. 


‘ Both the last stanzas,’ we are told, ‘form 


an almost independent picture, that has very 
little to do with the poem itself.’ After that 
there is little to be said, and there is only 
room for regret that a critic so acute as 
Herr Friedrich shows himself to be, and 
whose book is one of the ablest contributions 
to Horatian study made during recent years, 
should not see that whether an 
written for the Faunalia of December or 
February is a matter ludicrously unimport- 
ant, but that to say that in an ode of four 
stanzas two ‘have very little to do with 
the poem’ is to convict either the poet or 
the critic of folly. However, we must do 
him justice. He will have nothing to do 
with that famous conjecture of Bentley's 
1, 23, 5... vepris ad ventum, which a 
singularly able reviewer (Oxford Mag. 
October 31, 1894) has recently termed 
‘monstrous,’ and for this he deserves credit. 
But his defence of it is curious. ‘It might 
be objected,’ he says, ‘that the thought 
mobilibus veris inhorruit adventus foliis is 
too poetical for Horace. And yet we must 
not absolutely deny him genuine poetic 
feeling (eiyentliche Poesie), though it cer- 
tainly is shown least of all in the Odes.’ 
Poor Horace! Wisely was he in dread of 
pedagogues and professors. 


ode Was 


T. E. Pace. 





THE MANUSCRIPTS OF PROPERTIUS. 


(A DiscLaIMer.) 


OTHER engagements prevent me_ from 
replying to Mr. Housman’s article till the 
April number of the Classical Review. In 

the meantime I wish it to be understood 
that, neither as it stands nor with the 
unauthorized alterations of Mr. Housman, 


does the passage quoted on p. 27, col. 1, 
from p. 24 of my pamphlet, represent my 
views either now, or at the time of the 


amphlet’s publication. 
se J. P. PosrGare. 


ARCHAEOLOGY. 


PROF. CHRIST ON THE GREEK 
STAGE. 


Scenic archaeology has been much in- 
debted to Prof. Christ of late, not only for 
the three Munich dissertations which sprang 
from his interest in the stage question, but 
also for his own contributions to the subject. 


Beginning with a strong leaning stele 
the new theory because of its remarkable 
agreement with the requirements of the 
extant dramas, he has attempted in a — 
of investigations to bring order into _ 
chaos of apparently conflicting noWces, a 
to trace the various stages 1 the sre 
ment of the Greek theatre from Aeschylus 


134 


to Vitruvius. His first article (V. Jahrb. 
J. Phil. u. Paed. 1894, 27 ff.) was an attempt 
to define chronologically the successive 
changes in meaning of the words oxynvy, 
mpooKyviov, Noyetov, Oyuedy, Ke. It is to be 
regretted that he did not entirely succeed in 
throwing off the influence of Wieseler in his 
treatment of the subject. In a few points 
the prevailing belief has been corrected, but 
as a whole the work is in no sense final. 
The ground must be gone over again by 
one who is willing to know only what our 
sources teach us, if we are ever to be in 
a position to explain the statement of 
Vitruvius or to defend it—for he alone is 
our authority for a high stage in the Greek 
theatre. A second article in the Jahrbiicher 
(p. 157 ff.) is embodied in the third and 
most important, Das Theater des Polyklet in 
Epidauros in seiner litterar- und kunst- 
historischen Bedeutung (pp. 1—52 of the 
Transactions of the Munich Academy for 
1894). In this paper Prof. Christ appears 
as an opponent of the theory of Dr. Dirp- 
feld, and as the author of a theory of his 
own. As the first attempt at reconstruction 
offered by the conservative party, which 
frankly admits what the opposite side has 
proved, taking into consideration both the 
archaeological and literary evidence, this 
paper deserves careful consideration. 

The substance of the argument is as 
follows: The theatre at Epidauros is es- 
sentially in harmony with Vitruvius. The 
proscenium is a little further back and a 
little shallower than the Vitruvian, but its 
dimensions and position identify it un- 
questionably with the Vitruvian Aoyevov. 
Since Greek theatres of this type were still 
built in Vitruvius’ day and Greek plays 
still performed in them, it is impossible to 
reject the architect’s statement that this 
Aoyetov was the stage for actors. It cannot 
be explained as a GeoXoyeiov, because there 
was too little need of such a structure. The 
objections to this Aoyetov are that it is too 
narrow and too high. But it is not too 
narrow for three to four actors and not too 
high for the later drama; for the chorus 
had practically disappeared from the drama 
after the fourth century. The door in the 
proscenium proves nothing unless it can be 
shown that no door existed in the scena on 
the level of the Aoyeov. The proscenium 
door was useful for non-dramatic perform- 
ances. No platform was ever built up in 
front of it. When old tragedies were 
brought out on such a stage they were 
remodelled to suit the changed conditions. 
The Epidaurian proscenium could not have 


THE CLASSICAL 


REVIEW. 


served as the background of a classical 
piece because the central door was not large 
enough for the eccyclema. A high stage in 
the later Greek theatre is attested by 
vase-paintings, Athenaeus Mechanicus, and 
Pollux. 

The theatre at Epidauros, being unsuited 
to the classical drama, could not have been 
built by the elder Polycleitus. In uncon- 
scious agreement with Déorpfeld, Christ 
claims it for the younger Polycleitus, dating 
it ca. 300. What then was the form of the 
theatre of the fifth century? The oldest 
portions of the Athenian theatre are assigned 
by Dérpfeld to Lycurgus. But it is im- 
possible to believe that the Athenians were 
without a permanent theatre fer 150 years, 
and that this was provided only after the 
highest point of the drama was past. Ly- 
curgus simply enlarged the earlier theatre, 
making thirteen wedges instead of ten, and 
possibly extending the parascenia. But the 
scena-buildings of the fifth century were 
essentially the same as those which we 
assign to Lycurgus. 76 rpomvAaov 76 Avovicou 
Andoe. de Mys. 38, 7d texiov Arist. Leel. 
479, rydt wap’ aitiv tiv Oeov, Arist. Paw 725 
(statue of Athena), and zapackyvia Dem. 
Mid. 17, point to a stone theatre before 
Lycurgus. 

The parascenia of the oldest ruins at 
Athens are 20 m. apart and 5 m. deep. 
The dramas point to the use of the para- 
Scenia in scenes where one actor hides from 
others who stand in their regular position. 
The usual standpoint of actors was therefore 
in the space between the parascenia. That 
this place was elevated is proved by éxpiBas 
Plat. Symp. 194 B and by various passages 
in the dramas; an elevation for actors is 
indicated in Lys. 288 (7d cysdv), in Vesp. 
1342 and 1514, Hg. 149, Ach. 732, Eccl. 
1152 (dva- and xara-Baivev), and in Av. 175. 
This platform must have been 7—8 ft. high 
to admit of trap-door apparitions. An 
elevation in the orchestra is indicated by 
Eur. Here. 119, Hl. 489, and Jon 727 ff., 
and Av. 20 ff., 49 ff. Since chariots had to 
enter through the parodoi the platform in 
the orchestra was reached by inclined planes. 
Actors and chorus were in constant inter- 
course; therefore the difference in level 
between them was only about 2 ft., the 
connection being here also an inclined plane. 
Steps connected the lower platform with the 
auditorium. This huge double platform of 
120 sq. m., with its three inclined planes 
and its flight of steps, could have been no 
temporary structure, though built of wood. 
It remained from year to year unchanged, 





















although in exceptional cases the lower plat- 
form could be brought up to the level of 
the higher (thus spoiling the view from the 
front seats!). It came into use in the last 
decades of the fifth century. Before that 
time the chorus stood on the wooden floor 
of the orchestra, and the actors may have 
been on a slight elevation. 
A. multitude of objections to this mon- 
strous stage and thymele suggest themselves. 
It is a hopeful sign for the defenders of the 
new theory that Oehmichen and A. Miiller 
have so readily accepted it (W. Al. Phil. 
1894, July, and &. Ph. W. 1894, 1456). I 
shall confine myself to a few considerations 
which seem to me to be fatal to the two 
assumptions which form the basis of the 
‘whole argument: (1) that the Epidaurian 
proscenium could have been employed as a 
stage after the fourth century, and (2) that 
the plays of the fifth century could not have 
been performed in the orchestra in front. of 
this proscenium. At the outset be it re- 
marked that Christ gives only one explan- 
ation of the tremendous change in the 
height of the stage at the end of the fourth 
i century—harmony. He admits that it must 
f 
have been very awkward for actors to engage 
in conversaticn with the chorus from their 
12 ft. elevation. Miiller adds (/.c. 1459) 
that the occupants of the higher seats could 
not easily see over their neighbours’ heads 
what was going on in the orchestra, and 
that the very high stage was introduced 
for their benefit. But probably few others 
who have made the experiment at Epidauros 
would agree that an elevation was needed 
so far as the spectators were concerned. 
On the contrary, an actor standing between 
the proscenium and the centre of the 
orchestra could be better seen and better 
heard by the majority of the spectators than 
one who stood on the proscenium. 

‘The chorus had almost disappeared from 
comedy at the beginning of the third cen- 
tury. But there is strong reason for be- 
lieving that it continued to exist in tragedy 
far into the Christian era, and that its 
intimate relations with the actors were not 
essentially diminished! Although a char- 
acteristic tendency of the later tragedy was to 
make the songs of the chorus merely éPoApwa 
(Arist. Poet. 18), yet it is wrong to assert 
on this ground that the participation of the 
_ chorus in the action was thereby diminished 

(Christ, p. 26, ‘diese (Zwischenlieder) 


AM 


s 
’ 
s 


1 This is not the accepted view, but, I believe the 
correct one. A full statement of the evidence on 
this point will probably soon appear in the 
publications of the American School at Athens. 


THE CLASSICAL REVIEW. 135 


setzen ja keinen Wechselverkehr zwischen 
Chor und Biithne voraus’). Judging from 
the later plays of Euripides and by the 
thesus one might with more reason assume 


that, in proportion as the choral songs lost 
their vital connection with the plot, the 
greater became in compensation the import- 
ance of the chorus in the action. The 
function of the chorus in Roman tragedy, 
which was undoubtedly strongly influenced 
by the contemporary Greek tragedy, makes 
this more than probable. Satyric plays, in 
which the chorus always played an important 
part, continued down into the Roman period. 
There is no reason to believe that the 
classical pieces which were reproduced in 
Greece in the first century before Christ 
were emasculated by the omission of their 
choruses, while they flourished in almost 
their original form on the Roman stage. 
One cannot imagine that the Sophoclean 
trilogy with its Satyric after-piece, which 
was brought out in Rhodes in the first 
century, was performed without its choruses. 
Dion Chrysostom, it is true, speaks of the 
omission of the choral parts of tragedy 
(ca. 100 a.p.), but Welcker long ago has 
shown that this is not to be understood of 
the dramatic festivals in the principal cities 
even at that late period. The proscenium 
at Epidauros could not, therefore, have been 
an actors’ stage even after the fourth cen- 
tury. The vase-paintings have been re 
peatedly shown to have no bearing upon the 
stage in the Greek type of theatre. The 
notice of Pollux 4, 127, which implies a 
permanent connection between orchestra and 
stage, can also not refer to the Vitruvian 
Greek stage. The scaling ladders * such as 
are placed against the proscenia for the 
actors,’ mentioned by Ath. Mech. 29, are 
precisely such as the Paedagogue sets in 
position for Antigone in Phoen. 190. They 
mount from the street (rpios, v. 93), which 
passes in front of the palace, to the roof of 
the house, i.e. the top of the proscenium. 
The wooden ladder was not a permanent or 
a convenient arrangement, as the context 
ws. . 

he the other hand the plays of the fifth 
century are admirably adapted to the Epi- 
daurian theatre, provided that the actors 
place was in front of the proscenium. ee 
i inary, for the early proscen 

sine Pe ae Gene the doubted that the 
proscenium at Epidaurus 1s of much later 
date than the rest of the theatre. The top 
of the proscenium was not only ward 
Noyetov, but the roof of the house, which 


wood. 


136 


was often called into use. For a perfectly 
satisfactory explanation of the passages in 
the drama which indicate an elevation I 
need only refer to Dr. Bodensteiner’s ad- 
mirable discussion. Prof. Christ’s stage 
would have been impossible for the plays 
in which a tent occupies the background, 
of which several were produced after 420, 
and entirely unsuited to the majority of 
the plays brought out after this date. Ap- 
paritions occur in the plays of Aeschylus 
as well as in those of Euripides, and yet 
Christ makes no provision for the former. 
Too much stress is laid upon the existence 
of the parascenia for hiding scenes. It is 
noticeable that in all such scenes the person 
in hiding is able to see everything that goes 
on before the background (see esp. Ach. 
239 ff. and Herc. 1081 ff.). Therefore the 
actors did not stand back between the para- 
scenia. The only explanation that can be 
offered of the large parascenia in the early 
theatre is that they were used for the 
support of the wooden proscenium erected 
between them where the stone proscenium 
stood at a later period. , 

Prof. Christ’s paper is full of minor 
inaccuracies as regards the theatre ruins. 
He forgets that the floor of the early 
orchestra was of beaten clay, and that the 
mosaic pavement at Athens is of very late 
date. ‘The search for remains of a tunnel 
under this pavement, which he repeatedly 
urges, was made years ago without success. 
The number of xepxides in the theatre had 
nothing to do with the number of tribes. 
The round stone in the orchestra at Epi- 
daurus was certainly not the 6vpéeAyn, though 
it may have been connected with it. He 
is not aware that the ‘original five steps’ 
of the stage at Megalopolis were long ago 
given up by their own authors, or that 
fragments of probably the pre-Lycurgan 
cavea at Athens have been found. The 
door in the scena on a level with that in 
the centre of the proscenium, such as are 
found in several theatres, he failed to take 
into account when he erected his stage over 
it, 8 ft. above the dressing-rooms. One is 
surprised to find Mr. Haigh quoted in 
favour of a ‘thymele’ in the orchestra 6 ft. 
high, and Platonius cited to prove the 
default of the comic choregi in the fourth 
century, in the face of the explicit statement 
of Aristotle, ’A@. IloA. 56. His rejection of 
Dorpfeld’s dating of the Dionysiac theatre 
is hazardous, if no better grounds are given. 
The Romans got along without a permanent 
theatre for more than 150 years. It is 
noticeable that ai xvpiar éxxAnotar were 


THE CLASSICAL REVIEW. 


regularly held in the theatre after the time 
of Lycurgus, and not before. It is in- 
structive to note that the orators who spoke 
in these assemblies occupied, not the top of 
the proscenium, but the orchestra. The 
mpotvAaov To Avovicov of Andocides was 
not a part of the theatre, but the entrance 
to the Dionysus precinct. The re:xéov of the 
Ecclesiazusae was not necessarily of stone, nor 
the zapackyjvia mentioned by Demosthenes, 
though the latter may well have belonged 
to the structure completed under Lycurgus. 
The assumption of a statue of Athena on 
the proscenium from Pax 725 Wieseler has 
shown to be untenable. It is a noteworthy 
fact that Demosthenes, in the numerous 
passages in which he enumerates the great 
public buildings erected in the fifth century, 
never mertions the theatre. 
Epwarp Capps. 
The University. of Chicago. 


Description Raisonnée du Musée de St. Ger- 
main-en-Laye. Bronzes figurés de la Gaule 
Romaine, par SALOMON ReInAcH. Ouvrage 
accompagné d’une héliogravure et de 
600 dessins. Paris. Firmin-Didot. 1894. 
10 fres. 


Untit quite recently the Berlin Leschreibung 
der antiken Skulpturen was the only example 
of an official catalogue which aimed at fur- 
nishing illustrations of every object contained 
in the collection described. The catalogue 
now published by M..8. Reinach of the 
figured bronzes in the Musée de St. Germain 
is welcome as the first step towards the 
fulfilment of Professor Kekulé’s prophecy 
that the exception of Berlin would come in 
time to be the rule. M. Reinach not only 
reproduces 600 drawings (190 of which we 
owe to his own skilful pencil), but in a 
charming dedicatory letter to M. Bertrand 
gives a number of practical hints for the 
production of cheap yet adequate zinco- 
gravure. Every earnest student will be 
grateful to M. Reinach for the vigour with 
which he protests not only against non- 
illustrated catalogues, but against the eclectic 
method of illustration, and insists upon the 
necessity of representing the whole of a 
collection from its most important down to 
its apparently most insignificant object. 
For in the first place, as De Morgan long 
ago observed, no man’s selection of know- 
ledge is ever another man’s; and in the 
second, to quote our author, no one possesses 
an infallible criterion whereby to determine 
the value of objects, ‘telle statuette qui 





THE CLASSICAL REVIEW. 137 


narrétera pas un instant le regard d’un 
archéologue pourra fournir 4 un autre le 
point de départ d’une recherche fructueuse.’ 
Tf only compilers of catalogues would lay 
this to heart, what precious time would they 
not save students, by allowing them to know 
beforehand what to expect of a collection. 
An illustrated catalogue however cannot be 
more than a rough guide for those who have 
not yet seen a collection, or a convenient 
memorandum for those who have ; it cannot 
take the place of ocular examination or even 
_of photographs. It answers well for objects 
that have little artistic value and are inter- 
esting only as types, but it would be fatal 
to draw aesthetic conclusions from the 
caricatured reproduction for instance of the 
beautiful Blacas warrior (no. 182). 

But I am anticipating. Neither the 
actual catalogue nor the illustrations should 
cause one to overlook the Introductory 
Hssay in which, in his usual brilliant and 
incisive style, the author establishes the 
origin and character of Gallo-Roman art. 
In accordance with phenomena already 
partially indicated by Th. Schreiber, he 
recognizes that Greek influence made itself 
felt in Gaul mainly through the forms 
of Graeco-Egyptian art, which were im- 
ported not only through Italy, but directly 
wus a consequence of the relations between 
Graeco-Roman Egypt and the valley of the 
Rhone. Hence the numerous Alexandrine 
motifs—negroes, pygmies, children wrapped 
up in cloaks with pointed hoods, genre 
caricatures—that meet us at every turn in 
Gallic art. Hence also (a fact of the highest 
importance to the student of mythology) 
the types of the Gaulish gods. The type 
of ‘ Dispater’ M. Reinach clearly shows to 
be borrowed from Sarapis, while the curious 
crouching or squatting god whose origir has 
given rise to so much controversy he doubt- 
less rightly derives from the Egyptian 
Imhotep, in opposition to the current view 
which seeks his prototype in the crouching 
god of India. With regard to the whole of 
this complicated question we miss some 
reference to the labours of Prof. Rhys 
(Lectures on the Origin and Growth of Religion 
in Celtic Heathendom, 1888). In his searching 
analysis of the myth, Rhys, who makes 
express mention of the Cernwnnos of Autun 
(Reinach, no. 177), points out the connexion 
of the squatting god with the Norse Heimdal 
the ‘god’s warder,’ ‘whose teeth are of 
gold’ (op. cit. ‘the Gaulish Pantheon,’ 
pp. 78 sgq.). The gold teeth, symbol of 
metallic wealth, and the epithet ‘god’s 
warder’ seem to prove that we have in 


Heimdal and Cernunnos the survivors of at 
older race of gods displaced by 
Pantheon. The old gods are ret Cae, partly 
as servants of the new gods, while. like the 
dwarfs of the who are an 
instance of a similar survival, part of their 
functions is to toil in the bowels of the 
earth. 

The catalogue is agreeably interrupted by 
a number of excursus which take up points 
already touched upon in the Essay. 
Especially stimulating is the analysis of the 
type and the mythology of ‘Dispater,’ a 
full list of whose images is given in the 
alphabetical order of departments. The 
mallet or hammer, which is the most con- 
stant attribute of this god, Reinach explains 
on the supposition that ‘en général /es af- 
tributs des diewx sont des fétiches déchus’ 
hence that the diew au maillet was originally 
a dieu maillet, a mallet-god, on the analogy 
for instance of the axe-Dionysos at Pagasai, 
We wonder however that M. Reinach does 
not apply this illuminating theory to the 
Jupiter & la rove, no, 5. If the god with 
the hammer was the anthropomorphic de- 
velopment of the hammer-god, why should 
not the god with the wheel represent an 
original wheel-god? Certainly of the two 
explanations of the wheel by Gaidoz and 
Flouest respectively, quoted by M. Reinach, 
that of Flouest, who sees in the wheel a 
symbol of the thunder, comes nearest to the 
truth. This later symbolism gives the clue 
to the original meaning of the wheel ; its 
rumblings closely imitate those of the 
thunder ; hence it is a popular rain-charm ’ 
and a potent instrument of sympathetic 
magic. It may be further conjectured that 
the nine S-shaped objects which the Jupiter 
carries on a ring slung over his right 
shoulder were also instrumental in imitating 
the rattle of the storm. The mallet was 
doubtless part of the same artillery (cf 
however Rhys, Joc. cit. p. 67). | 

At every turn of the catalogue interesting 
points connected with cult and symbolism 
suggest themselves: thus the numerous 
votive horses, no. 296 sqq., recall our own 
‘white horse’ in Berkshire, commemorative 
in popular tradition of a battle between 
Alfred and the Danes. But the student of 
art, and of Greek art, will also tind the 
catalogue full of valuable material. I need 

1f J. G. Frazer, Golden Bough, vol. i. p. 21. 
Mr. Frazer’s explanation of the chariot of Krannon, 
and his remarks on rain charms in general had 
escaped my memory, both when ! wrote ie 
CLR, 1894, p. 175, and when I was preparing bese 
English ed. of Furtwiingler’s book (cf. Masterpieces 


of Gree > Sculpture, pp- 469 sqq.). 


hew 


Niebelunqgen ~ 


138 THE CLASSICAL 


only call attention der alia to the charming 
Hermaphrodite, no, 118, reminiscent of the 
Kallipygos, as the author points out, and 
still more of the graceful Satyr looking 
back at his tail (Vatican, Helbig 371); to 
the Satyr drockorevwy, no. 113, a bronze of 
capital importance for the history of art, forit 
is doubtless an echo of the Satyr of Anti- 
philos ; to the lovely Hypnos, no. 102, after 
a celebrated Greek original. A fine bust, no. 
213, M. Reinach well assigns to the group 
of Polykleitan heads represented by the 
Doryphoros of Apollonios in Naples. Most 
superb of all however is a head with turret 
crown, no. 91, described as Génie de Ville ou 
Cybele (Cab. des Meédailles), Caylus was 
most certainly right in refusing to see in 
this head a Gaulish work; he suggested 
that it might have been brought from Rome 
to Paris ; its ultimate provenance seems to 
me more likely to be Greece. It can at 
any rate have nothing to do with the 
Renascence ; I should suggest that it was 
copied from a fine Greek original ; the fea- 
tures have the breadth and the seriousness of 
the fifth century, while the coiffure appears 
with slight variations on a number of 
Greek heads of that epoch (with single 
band, Berlin Skulpt. 608, and J.H.S. ix. pl. 
IV., with double band, Rim. Mitth. Taf. 
vii. p. 165, fig. 1, etc.). As tothe mural 
crown, which represents a mediaeval castle, 
the lack of organic connexion between 
it and the head shows it to be a_ late 
barbarous addition (cf. the ring on the 
head of the fine centaur from Spires, 
no. 117). A charming section is formed by 
the vases decorated with reliefs (nos. 394— 
434), representing scenes for the greater 
part homelike or idyllic which so clearly 
attest Alexandrine influence. No. 414 
should be especially noted as affording one 
of the instances, so rare in classical art, of 
Orpheus playing to the animals. 

We must perforce admit with M. Reinach 
that an ‘admiration de commande’ for 
Gaulish works would be quite out of place. 
Yet, clumsy and unintelligent though the 
Gauls were as artists or handicraftsmen, it 
must be owned that they showed themselves 


appreciative of good work by importing 


such beautiful objects as the Blacas warrior 
or the turreted head mentioned above, and 
by not unfrequently choosing good models 
to copy or to adapt. The result is that the 
classical archaeologist can no longer afford 
to neglect Gaulish art. When he does turn 
to it he will find no more stimulating or 
suggestive introduction to the whole sub- 
ject than M. Reinach’s admirable De- 


REVIEW. 


scription Raisonnée of the Gaulish figured 
bronzes. 
Eugenie SELLERS. 


A. WALTON ON THE CULT OF 
ASKLEPIOS. 


Cornell Studies in Classical Philology. No. 
IlJ. The Cult of Asklepios, by ALtcE 
Watton, Ph.D. 1894. 


APPEARING as it does under the auspices of 
a learned society, this is a disappointing 
treatise. Dr. Alice Walton states quite 
frankly in her preface that the monograph 
is a compilation, ‘It has been my aim to 
give in narrative form the results obtained 
by a careful comparison of material from 
the different localities, and also to show by 
means of indexes what material is used.’ 
The indexes of localities, literature, inscrip- 
tions, etc., occupy considerably more than a 
third of the book, and are finished to a 
perfection truly American, the text itself is 
abundantly learned, and yet clear and read- 
able, but there the matter ends. Von 
Wilamowitz has it seems said the last word 
so far on the intricate cult of the me- 
dicine god. Dr. Alice Walton’s popular 
compilation adds no single original sug- 
gestion. It will nevertheless be of high 
use to students. On p. 28 is a discussion of 
the familiar reliefs with the enigmatic head 
of a horse in an ineuse square. Dr. Walton 
rightly rejects Le Bas’ theory that this 
represents the ‘steed of Thanatos.’ She 
accepts, in lieu of a better, the scarcely 
more satisfactory theory that the horse 
recalls the ancient custom of burying 
favourite animals with their master. Is 
not a third solution possible? Asklepios 
came from Thessaly, from Trikka, the 
ancient home of the horse-god, peopled by 
the cavalry race figured mythologically 
by the Centaurs; he was himself nurtured 
by a Centaur—was he not at one timea 
Centaur himself, Asklepios Hippios, and is 
not the horse’s head the surviving symbol, 
the crest as it were of the ancient half- 
forgotten cult ? 
JANE HE. Harrison, 


MONTHLY RECORD. 


ITALY. 


Ancona.—In the piazza Cavour have been dis- 
covered six tombs dating from the third century B.c. 
to Roman times, and showing that the necropolis 
continued on this site until the latter period. The 


ne 


i 


























































remains of several walls, important for the topogra- 
phy of the city, have also been found.! 


THE CLASSICAL REVIEW. 139 
are filled by two amphorae. Most of the design is is 
black outline, with details in green, blue, & Tha 


Arezz0.—On the site of the potteries Anniv, Mem- 
mia, Rasinia, have been found a number of terra: 
cotta fragments belonging to the upper part of a 
small temple. They include a fragmentary bas- 
relief with traces of colour, representing a Nereid on 
a maripe monster, carrying a greave (?), probably 
part of a frieze representing Nereids carrying the 
arms of Achilles ; an acroterion with the head of a 
nymph ; fragments of dmbrices, &c.? 

Caltrano Vieentino (Venetia).—A hoard of 365 
victoriati, more than half of which belong to the 
period 228—-217-B.c., has been found. ! 

Colle del Vallone, near Paganica.—Several tombs 
have been opened, one containing a cinerary urn, 
the others having served for inhumation. The bodies 
were placed either on the bare earth, or in wooden 
coffins, of which the nails and metal corner-pieces 
only remain. No tomb-stones of any kind were 
found, nor any coins or objects in bronze, whereas 
fictile vessels and iron objects were numerous. ! 

Montemarciano (Umbria),—A hoard of 208 Re- 
publican denarii has been discovered. 

Montepulciano (Etruria).—The following objects 
have been found in a sepulchral chamber: a bronze 
kottabos-stand, 1°30 m. high, with broken iron base, 
and surmounted by a small figure of the Etruscan 
Charun or Tuchulcha ; a pair of bronze candelabra 
capped by a group of a man bridling a horse ; several 
bronze vessels ; an iron and bronze brazier on wheels ; 
fragments of an iron spade; and the bottom of a 
kylix of Orvietan fabric, which dates the tomb about 
the end of the fourth century B.c.! 

. Quatrelle (Venetia).—A bronze Roman weight of 
101°30 grammes, with an inscription EX CA 
ex[actum ad] Ca[storis], has been found in a tomb of 
the Antonine period. Among the objects with it 
were a leaden weight of 103 gr., early Imperial 
coins (including one of Antoninus Pius struck for the 
Lycopolite Nome), some glass, and a singular object 
(possibly a priapiform lamp).* 

Rome.—Reg. Il]. A fragment of an ancient cal- 
endar (part of September and October) has been 
found, in the course of the continuation of the via 
de’ Serpenti. It belongs to the early years of the 
reign of Tiberius. Among the festivals mentioned 
are the Fontinalia: [feriae] FonTI ExTRA P(orlam 
Capenam *).+ 

Reg. VII. In the via Capo le Case has been found 
a fine male torso, in marble, slightly above life- 
size. 

S. Angelo in Formis.—A large tile has been dug 
up, bearing a graffito inscription which reads as 
follows: N.D.E.c| Idibus Lulis Celer finget | bi- 
pedas ¥Xxx1 | Actum Casilino | Modesto 11 et Probo 
cos|... It is a contract for the making of 5031 


tiles by a certain date. The date of the document is 


228 a.p. The first four letters may stand for n(onis) 


 dee(embribus) or for n(ominc) d(ccwrionwm) e(t) e(olon- 


orum),” 
Sorrento.—A milestone with the number XXv., 
and dedicated to the Emperor Maxentius, has been 
found. It belongs to the road from the promontory 
of Minerva to Pompei.* 
Taranto. — Three mosaic pavements have been 
discovered, one of which—the only one in Taranto 


_ with a representation of the human figure—repre- 


sents Bacchus, nude, standing to front ; with his 
left hand he leans on a thyrsus, in his right is a 
cantharus ; at his feet a panther. The upper corners 


1 Notizie degli Scuvi, July 1894. 


2 Ibid. August 1894. 
3 Ibid. September 1894. 


frame of the picture is a row of rectangles (double at 
the top and bottom) containing rhomboids in which 
again are inscribed small circles. Lowest of all is a 
band divided into three parts ; the midds me cor 
tains three very rudely designed busts; the others 
are filled each by a double rectangle with semi-cireles 
arranged along the inner sides, a dise in the middle, 
and lines dividing the field into six small r 
angles.3 

Verona,—Excavations on the site of the amphi 
theatre have made clear many details of the building, 
including especially the course of the subterranean 
channels, and the substructions of the seats.' 

Verucchio and Spadarolo near Rimini.—An ex- 
tensive cemetery of the Villanova type has been 
systematically examined. It contains a group of 
very archaic tombs. The cinerary urns are with one 
exception of very rude fabric, of the shape of a 
double truncated cone, with lid, and a single handle. 
They have bands of shallow incised geometrical 
decoration. The tombs were placed close together, 
and in a space barely 3x 4°50 m. were fonnd some 
thirty tombs, the urns standing one on top of 
another. In one tomb a skeleton was found, 
elsewhere the bodies had been burnt. The objects 
found were mostly of bronze ( jidulae, armlets, ete.). 
Among the fictile vessels were (tomb 10) a two- 
handled vase of black clay, the handles of which 
were decorated with a row of small whitish circles 
produced by inlaying rings of shell or bone. Tombs 
21 and 46 contained iron daggers. In 52 was a 
fibula, the spur of which is curved, and has an amber 
head, while the cord of the bow carries a quadrangu- 
lar frame with two small horns at one end. In the 
same tomb was a careful though primitive clay copy of 
a double-crested helmet similar to the bronze helmets 
from Tarquinii, with rows of impressed circles 
alternating with incised lines and rows of triangles, 
At Spadarolo, among several objects of a similar 
nature, was found an open-work bronze ornament 
(2), representing a human figure between two birds ; 
it is of the type frequently found in Etruria, and 
evidently developed out of the Mycenaean style of 
ornament described by Evans (J. H.S. vol. xi. pp. 
197 ff.). This specimen however stands on a support 
with a rectangular base, to which is fastened by the 
hands a small figure. It is suggested that the upper 
part formed the handle, and the lower a support, of 
a hemispherical cup, the point of attachment being 
the rectangular base already mentioned, which is 
slightly curved.* 

GREECE. 


Athens.—A small relief of Asklepios, surrounded 
by four sick persons, has been discovered, together 
with a mosaic perement on which is represented a 
hippalektryon.* In the theatre of Dionysus, the 
subterranean passage by which the actors gained 
access to the stage, has been found.’ The whole of 
the ground-plan of the temple of Dionysus in Limnae 
has been laid bare.® ; 

Kalauria.—Excavations on this island have re 
sulted in the discovery of the peribolos of the already 
known Ionic temple. It is 56 metres long by 28 
broad. The wall consists of unworked blocks of 
dark limestone and poros stone, A propylacum 
adorned each of the two entrances to the enclosure, 
Both temple and peribolos belong to the 6th ae 
p.c. A further discovery is that of a stoa built o 


4 Athenaeum, February 9. 
5 Standard, February 14. 
® Athenacum, February 16. 


140 


polygonal masonry, apparently belonging to the 
second part of the 5th century B.c. ; while another 
stoa is supposed to have been built by Eumenes II. 
A propylaeum leading to the square before the 
temple, a stoa to the west of this propylaeum, and a 
courtyard surrounded by small rooms are among the 
other buildings discovered.? 

Crete.—Halbherr’s explorations during the last 
year have covered two-thirds of the eastern part of 
Crete. Besides vases, ranging in style from the 
Theran to the archaic Greek types he has discovered 
a number of terra-cotta statuettes (some with a 
peculiar stamp attributed by the finder to the 


7 Athenaeum, January 19. 


THE CLASSICAL REVIEW. 


Eteocretans), and of very archaic steatite stones, 
incised in some cases with marks bearing on the new 
theory of a prehistoric script. As to the question of 
interment in Mycenaean times, the evidence of 
Erganos is in favour of inhumation, that of other 
places in favour of burning. Some important sculp- 
ture has been found, including a good relief of a 
dancing girl. There are several inscriptions of 
importance, both archaic and of later times, including 
one dated in the reign of Demetrins Poliorcetes, and 
a Latin reseript.® 


G,. F. Hitt. 


8 Academy, January 19, 





SUMMARIES OF PERIODICALS. 


Journal of Philology, Vol. xxiii. No. 45. 


1894. 


Excerpts from Culex in the Escorial MS., Robinson 
Ellis. Further Suggestions on the Actna, Robinson 
Ellis. Tdese include notes on the Stabulensian 
fragment in the National Library of Paris [see 
Journ. Phil. xvi. 292]. On Herodas, Robinson 
Ellis. All arguments in favour of the Alexandrian 
epoch of Herodas are more or less unconvincing. 
Prof, Ellis thinks it not inconceivable that Herodas 
lived later than Catullus and Vergil [see Cl, ftev. v. 
457]. Critical notes on various passages are also 
given. Did Augustus create Hight new Legions 
during the Pannonian Rising of 69 A.v.%, E.G. 
Hardy. Maintains, in opposition to Mommsen, that 
only four new Legions were raised at this time, and 
that before the Rising the Roman army numbered 
at least twenty-two Legions. Thucydides and the 
Sicilian Expedition, W. E. Heitland. Taking Thue. 
as our one first-class authority for the history, Mr. 
Heitland attempts to clear up some obscurities [see 
Cl. Rev. i. 73 and viii. 123]. Plato, Phaedo ch. 48, 
Colin E. Campbell. A criticism on Mr. Archer- 
Hind’s interpretation of. this chapter. #8) and 3q 
in Homer, ¥. W. Thomas. Maintains that these 
particles are primarily temporal and refer to some 
new or critical event just occurring. The only 
difference between them is that #5 ismore emphatic, 
and, as containing 7, is almost restricted to the 
speeches. On the Text of M. Aurel. Antoninus 70, 
eis éavtdv, Gerald H. Rendall. Considers that the 
text of the Meditations is ‘singularly susceptible of 
secure emendation,’ of which many specimens are 
given. : 

American Journal of Whole 
No. 59. October 1894. 


William Dwight Whitney, by T. D. Seymour. 
In his death ‘this country has lost cne of her most 
distinguished men, one who had been recognized 
throughout the world as one of the highest author- 
ities in his department of learning, and who had 
been for forty years the leader of oriental and lin- 
guistic studies in America and the personal master 
of a majority of the American scholars in his de- 
partment.’ The Latin Prohibitive, part ii., by Hie: 
Elmer. This paper is chiefly devoted to the use of 
neque (nec) with the perf. subj., and seeks to show 
that this construction does not occur as a prohibition 
(outside of poetry) till the beginning of the period 
of decline, that the few w/-clauses in Cic. continued 
by nec, supposed to be purpose-clauses, are really 


Philolopy. 


result-clauses, while other cases of nee with the pert. — 
subj. are really examples of the (so-called) potential 
subj. and to be translated by ‘would’ or ‘should.’ 
The Judaco-German Element in the German Lan- 
guage, by L. Wiener. The Notes consist of Cor- 
rections and Additions to Lewis and Short, by F. G. 
Moore and Brief Notes on Plautus Terence and 
Horace, by A. F. West. The passages noted are 
Pl. Rud. 489—90, Trin. 512, Ter. Heant. 342, Hor. 
Od. i. 17,.20, Sat, i1..6,°79, Epp<i,.1; 2-anda, 772! 
The only review devoted to classical philology is 
of the two new Pitt Press editions of Plautus— 
Stichus by Dr. Fennell, and Epidicus by J. H. Gray. 
The reviewer, E. W. Fay, proceeds from the con- 
servative standpoint that the MSS. are a better guide 
than metrical theories. 


Jahresberichte des Philologischen Vereins 
zu Berlin. January—May 1894, (Continued from 
p- 96). 

ON THE LITERATURE OF ConNELIUS NEpPos, by 

G. Gemss. 


I. Editions. Corneli Nepotis Vitae, by W. 
Martens. 3rd edition. Gotha, 1893. Shows much 
improvement on the last edition and is up to date. 
Cornelii Nepotis Vilae, by M. Gitlbauer. Ed. 4. 
Freiburg- i.-B. 1893. Practically the same as the 
3rd edition, on which see Classical Review vii. 383. 

II. Contributions to criticism and elucidation. 
Lange, N. Jahrb. f. Phil. Some interpretations and 
emendations. H. Muzik, Bemerkungen zu Weidners 
Neposausgabe. Pr. Krems 1892. Seeks to show 
that W. has often emended unnecessarily. Next 
follow some criticisms by the reviewer on Atticus 13, 
2 and 3, 2. Vahlen in Herm. xxviii. defends the 
MS. omniwm in Epam. 1, 4 where animz is generally 
read. 

Il]. Appreciation of Cornelius Nepos. 4G. 
Daichendt, Die Lektiire des Cornelius Nepos. Pry. 
Bistritz 1890. This has especial reference to the 
Christian standpoint. J. Weissenborn, Cornelius 
Nepos in seiner Bedeutung fiir den Unterricht. Pr. 
Aschaffenburg 1892. An excellent work that may 
be recommended to all. 

IV. On the text. H. Muzik has found a MS. of 
Nepos of the 15th cent. at Gottwei. A collation of 
the first four lives shows that it belongs to the class 
MR. ‘Traube, Untersuchungen zur Ueberlieferungs- 
geschichte romischer Schriftsteller, Sitzungsber. der 
philos.-philol.-historischen Kl, der k. bayer. Ak. d. 
Wiss. 1891. Probus never composed anything but 
only epitomized existing lives, 


many Berc?? 


















THE CLASSICAL REVIEW, l4] 


V. Dissertations. F. Fiigner, Des Cornelius Nepos 
Lebensbeschreibungen. Leipzig 1893. Part L., Text. 
Part II., Notes. This is the first vol. of school 
editions of Greek and Latin authors under the 
supervision of Fiigner. Here the notes are rather 
too long and the vocabulary too short, though it is 
an excellent edition on the whole. Cornelius Nepos. 
Selected Lives, by P. Ditsch. Leipzig 1894. Part 
I., Text. Part II., Notes. This also is part of a 
collection of Greek and Latin authors. It contains 
nothing that is not important for school use. 


ON THE LITERATURE oF Livy, by H. J. Miiller. 


I. Editions. 7. Livia. u. ec. libri. 
edition, by H. J. Miiller. Vol. i. Part IJ. (Bk. 
2). 8thedition. Berlin 1894. Has been subjected 
to a thorough revision, and more attention has 
rightly been given to the sense and use of language 
than to palaeographical matters. 7. Liviia. u. ec. 
liber xxii., by F. Luterbacher. 3rd edition. Gotha 
1894. This excellent edition has been carefully 
revised. C. Haupt, Livius-Kommentar. Books vi. 
vii. and (separately) book xxii. Leipzig 1893. 
Much to be commended, though there are traces of 
haste. TJ. Livii a. u. c. liber xxix., by F. Luter- 
bacher. Leipzig 1893. Maintains the high standard 
of Ix’s work: 7. Livi a. u. c. libri. Ed. A. 
Zingerle. Pars vi. Fase. I. xxxvi.—xxxviii. Prag 
1893, and A. Zingerle, Zur vierten Dekade des Livius. 
Wien 1893. The former is a careful.and solid work, 
the latter (a dissertation) justifies Z.’s conjectures 
and emendations. W. Vollbrecht, Auswahl aus 
Livius xxi.—xxx. Leipzig 1893. An excellent 
seleetion, which might better be called ‘The Hanni- 
balian War.’ H. Geist, Was bieten die antiken 
Historiker der imodernen Jugend? Posen 1891. 
An admirable treatise which shows how the spirit of 
antiquity constantly renews humanity. G. Hergel, 
Klassikerlektiire und Realien. Zur Livius lektiire. 
Pr. Briix 1892. Recommends for the study of 
Realien the original writers and not commentaries 
on them. 

II. Contributions to criticism and elucidation. 
O. Keller, Zu Livius. Zeitschr. f. d. 6st. Gymn. 
1893. Contains five emendations. R. Bitschofsky, 
Kleine Beitrage zwr Kritik und Erklérung einicer 
Stellen des Livius. Evanos Vindobonensis (Wien 
1893). Critical notes on 2, 30, 1, 2, 36, 3 and 22, 
31, 5. B. Kruezkiewiez, Livianwm. Cracoviae 
1893. On 1, 21, 4 defending the text soli Fidei 
sollemne “instituisse [institutt 2]. Scattered contri- 
butions are found as follows : E. Meyer (Pr. Herford 
1893) on 1, 14, 8, C. v. Morawski (Zeitschr. f. d. 
ost. Gymn. 1893) on 9, 19, 6, F. Walter (Bl. f. d. 
GSW. 1893) on 9, 33, 3, W. Heraeus (WS. f. klass. 
Phil. 1893) on 10, 14, 18, F. Luterbacher (N. Jahr. 
f. Phil. 1894) on 21, 37, 4 the fifteen days of 
Hannibal’s passage of the Alps, J. B. Greenough 
(Harvard Studies iii. 181) on 22, 17, 2, M. Miiller 
(br. M.) on 22, 42, 12, 51, 5, 55, 8 and 58, 7, F. 
Fiigner (Berl. Phil, WS. 1893) on 30, 25, 6 and 29, 
4, E. Wolfflin (Archiv f. lat. Lex. 1892) on 36, 15, 
4, F. Luterbacher (N. Jahrb. f. Phil. 1893) on 37, 
56, 2, C. Funck (Archiv f. Lat. Lex. 1892) on 44, 
81, 1 defends wtrarios against Madvig’s putearios. 

III. Lexicon, Sources, &c. Lexicon Livianwm, 
conf. F, Fiigner. Fasc. vi. ambitio—annuus. Comp. 
F. Schmidt. Leipzig 1894. R. Becker, Bildnisse 
der Geschichtschreibers Livius. Leipzig 1890. Chiefly 
devoted to a description of a marble bust at Padua. 
C. v. Morawski, Zur Rhetorik bei den riimischen 
Historikern (Livius, Velleius, Curtius). Zeitschr. 
f. d. dst. Gymn. 18938. W. Soltau, Die annalis- 
tischen Quellen in Livius’ vierter und tiinfler Dekade. 


Weissenborn’s 


Philol. 52. S. maintains that Livy combined t 

sources of information (1) the pontifical anna 
through Piso and Valerius Antias, (2) for Greek 
eastern events Polybius and Claudius. A. Volkmar, 
De annalibus Romani quaestiones, I. de historia 
decemviratus. II. De T. Livio fonte Dionysi Hali- 
carnassei. Diss. Marburg 1890. V. seeks to sho 
in I. that L. has represented Caesar under the person 
of Appius Claudius. J. Schell, De Sulpicio Seve 

Sallustianae, Livianae, Taciteae clocutionis tm itaten r 
Diss. Miinster 1892. W. Boguth, Markus Valeri 

Livinus, Pr. Krems 1892. A most praiseworthy 
contribution to the history of the Second Punie War. 
J. Fuchs, Der zweite Punische Krieg and its sources 
Polybius and Livius examined from a g 

tactical standpoint. Shows thorough research and 
extensive knowledge, but the writer has attempted 
more than can be proved. 

Addition. TZ. Livii a. u. e. liber xxi., by F. 
Luterbacher. 4th edition. Gotha 1894. The early 
appearance of a new edition is a testimony to its 
worth. 


iow 


Strutevgic- 


Rheinisches Museum, Vol. xlix. l'art 4. 
1894. 

Zwei neue Reden des Choricius, R. Forster. These 
two speeches are here first published from a Madrid 
MS. Harpalyke, G. Knaack. Confirms O. Crusius 
rewark (in Roscher’s Mythological Lexicon) that 
Camilla is a Roman copy of Harpalyke, and refers 
to Servius and Hyginus. <Ancedota medica Gracea, 
R. Fuchs. I. List of contents of cod. Paris. 
supplem. Grace. 636. II. Collation of fol. 102 v.— 
105 v. the Canon of Maximus Planudes. JIL 
Inedita medica from pp. 21—S82. <Autor- und Ver- 
lagsrecht in Alterthum, K. Dziatzko. Concludes 
that these did not exist among the Greeks and 
Romans. Zui Datirung des Delphischen Pacan une 
der Apollo-Hymnen, H. Pomtow. The Paean 230— 
220 3B.c., the four hymns at different times 185— 
135 3.c. Das Regenwunder der Marc-Aurel-Sdule, 
A. v. Domaszewski. Seeks to show that the evidence 
of this column is against the genuineness of the 
Christian legend of the ‘thundering legion.’ 
MISCELLANEOUS. Zu Aeschylos Choephoren, J. M. 
Stahl. A redistribution of ll. 498~-511. Theopomp, 
E. Rohde. On a letter of Alexander the Great to 
the Chians. Parthenius, FE. Rohde. On a passage 
at the end of ch. 36. TAg@aoa, H. Rabe. From 
cod. Mare. gr. 433. Zur Datirung der Halle der 
Athener zu Delphi, H. Pomtow. B.C. 490 is the 
terminus ante quem. Zu Martial ii. 17, Ch, Hulsen. 
Die Gallischen Steuern bei Ammian, O. Seeck, On 
Amm, Mare. xvi. 5, 14. 


Neue Jahrbiicher fur Philologie und Faeda- 
gogik. Vol. 149. Parts 10,11. 1594. 

Fasti Delphici II, 1, H. Pomtow. This (a con- 
cluding article) is on the Amphictyonic decrees of 
the second century B.c. with a revision of the text 
and four appendices. Kritische bemerkungen 0 
Xenophons Kyrupidie, K. Lincke, Among the 
passages discussed is the long one vil. 5, $$ 2 -16, 
and v. 2, §§ 16—19 on the supper in the Persian 
camp. Lu Nenophons aera ey | Pr. ht, 

iiller. Suggests # oxvov in 1. 4, or } wévow, 
racine ii. 6, 1 Zur Eudemischen chik, QO. 
Apelt. On various passages in the seventh and 
eighth books, from 1235"—1245*. Also a few pass- 
ages in the first book are noted. Zur bean lung 
des Sapphischen maszes bei Horatius, R. Kipke. 
From the fact that the fem, caesura in the rg is 
found in 25% of the sapphic lines in the fourth book 
of the odes and the carm. saec,, and only in 145% of 
those lines in the first three books, it is concluded 


143 THE CLASSICAL REVIEW. 


that Horace deliberately in the last book (and in i, 
30 by way of experiment) emancipated himself from 
the Roman school-theory which required the masce. 
caesura, into the greater freedom of his Greek models. 
Zu Caesars bellum Gallicum, G. Hubo. Ini. 52, 4 
reiectis pilis comminus gladiis pugnatwm est, for 
yeiectis reads relictis the reading of most MSS. 
Studien zur tiberlieferung und Kritik der metamor- 
phosen Ovids, H. Magnus. IV. Marcianus and 
Neapolitanus 2. Der Bubastische Nilarm, W. 
Schwarz. Greek writers who did not live in Egypt 
(as Diodorus) spoke of a Pelusiac branch of the Nile, 
those who lived in Egypt (as Manetho und Ptole- 
maeus) spoke of a Bubastic branch and a Pelusiac 
mouth. 


Mnemosyne, N.S. Vol. xxii. Part 4. 

De Theogonia Orphica (concluded), A. KE. J. Hol- 
werda. ‘The passage of Damascius does not by 
itself prove that an Orphic @eoAoyia beginning with 
Nvé ever existed. Yet scholars join the Orphic 
verses found in Plato, taking them from the 
paywdia to which they beling (4 ouv7@ns oppexh 
OeoAoyla 7 év pay@diats says Damasc.), with the Nvé 


of Aristotle or Eudemus with which they have no 
connexion. <Analecta critica, L. K. Enthoven. 
Notes on passages of Appian, Artemidorus, Dionys. 
Halic., Dio Chrysost., Herodian, Plutarch, Them- 
istocl. epist., and Zosimus. Observationes eriticac ad 
epistolographos Graecos, G. M. Sakorraphus. The 
result of the examination of some MSS. at Vienna 
and some Italian cities. Notes on the letters of 
Aristaenetus (Cod. Vindob. 310). A text is given of 
the témo. emucroArkot of Demetrius Phalereus by the 
aid of Cod, Marcianus 418, and the variations are 
noted from Hecker’s ed. Cie. de prov. cons. § 33, 
J. v. d. V. Varroniana, I. C. G. Boot. Passages in 
the De lingua latina considered. Ad Cie. orat. pro 
Clucntio annotatiunculae criticae, J. J. Hartman. 
Observatiunculae de ture Romano (continued), J. C. 
Naber. This treats. of De condictione propter 
poenitentiam, de possessionis condictione, de con- 
dictione fructuum, and de constitutione ad Aufidium 
Victorinum. Ad Homeri Iliadem, H.v. H. On 
Z 335 suggests veweotu? for veuéoor and on I 707 
modAn for gary. Valerius Maximus (ii, 1, 8), 
Javed Vs 
Re Css: 





LIST OF NEW BOOKS. 


ENGLISH BOOKS, 


Aristotle. Politics. A revised text, with intro- 
duction, analysis and commentary by Franz Suse- 


mihl and R. D. Hicks. Books I.—V. 8vo. xv, 
689 pp. Macmillan & Co. 18s. net. 
Aristotle’s Theory of Poetry and Fine Art. With a 


critical text and a translation of The Poetics, by 
S. H. Butcher. 8vo. xix, 384 pp. Macmillan 
& Co. 10s. net. 

Burton (E. de W.) Syntax of the Moods and Tenses 
in New Testament Greek. 2nd edition, revised 
and enlarged. 8vo. 220 pp. Clark. 5s. 6d. 
net. 

Collar (W. C.) and Daniell (M. G.) The First Latin 
Book. 12mo. xiii, 286 pp. Boston, Ginn. 
Cloth. Illustrated. $1. 10. 

Cornelius Nepos, Miltiades, Themistocles, Aristides, 
Pausanias, Cimon, with notes and vocabulary for 
beginners, by E. 8S. Shuckburgh. Fep. 8vo, ix, 
86 pp. Cambridge Press. 1s. 6d. 

Cornell Studies in Classical Philology. No. III. 
The Cult of Asklepios, by Alice Walton. 8vo. 
Boards. viii, 136 pp. $1. 25. 

Dawes (EK. A. S.) The Pronunciation of the Greck 
Aspirates. 8vo. 103 pp. Sewed. Nutt. 2s. net. 

Furtwangler (A.) Masterpieces of Greek Sculpture. 
A series of essays on the history of art. Idited 
by Eugénie Sellers. With 19 full plates and 200 
text illustrations. Folio. 486 pp. Heinemann. 
63s. net. 

Gleason. The Gate to the Anabasis. With collo- 
‘quia, notes and vocabulary. By Clarence W. 
Gleason, Master in the Roxbury Latin School, 
Boston. ‘School Classics Series.” 16mo._ iv, 
107 pp. Boston, Ginn. Cloth. 45 cents. 

Hammond (B. E.) The Political Institutions of 
the ancient Greeks. 8vo. Cambridge Press. 4s. 

Homer. Jliad. Book XXIV. With introduction, 
notes and appendices, by G. M. Edwards. Feap. 
8vo. xxxvi, 74 pp. Cambridge Press, 2s, 

—— Odyssey. Books V.—VIII. Edited with 


introduction and notes by B. Perrin, Professor of 
Greek in Yale University. ‘College Series of 
Greek Authors.” Square 12mo. iv, 186 pp. 
Boston, Ginn. Cloth. $1. 50. Text Edition. 
62 pp. 45 cents. 

Horace. Odes. Books 1 and 2 done into English 
verse, and Andromeda, Ariadne, and Jason, by J. 
H. Deazeley. Impl. 16mo. Clarendon Press. 
7s. Gd. net, 

— The Historical and Political Odes. 
duction and notes, by A. J. Church. 
146 pp. Blackie and Son. 2s. 6d. 

Floratius. Odes and Epodes. Edited with intro- 
duction and notes by Clement L. Smith, Professor 
of Latin in Harvard University. ‘College Series 
of Latin Authors.’ 12mo. Ixxxvii, 404 pp. 
Boston, Ginn. Cloth. $1. 60. Text Edition. 
45 cents, 

Lindsay (W.M.) The Latin Language : an historica 
account of Latin sounds, stems, and flexions. 
S8vo. 660 pp. Clarendon Press. 21s. 

Lueian’s True History. Translated by Francis 
Hickes: illustrated by William Strang, J. B. 
Clark and Audrey Beardsley : with an introduction 
by Charles Whibley, 4to. 270 pp. Privately 
printed. 42s, net. 

Lord (F. E.) The Roman Pronunciation of Latin. 
Why we use it and How to use it. 12mo._ iv, 58 
pp. Boston, Ginn. Cloth. 40 cents. 

Pater (Walter.) Greck Studies, a series of Essays 
prepaed for the Press by Charles L, Shadwell. 
8vo. ix, 315 pp. Macmillan & Co. 10s. 6d. 

Perrot (G.) and Chipiez (C.) History of Art in 

« Primitive Greece. Mycenian Art from the French, 
illustrated, with 544 engravings in the text, and 
20 coloured plates. 2 Vols. Roy. 8vo. 1046 pp. 
Chapman, 42s. 

Plato. Republic. The Greek text. Edited, with 
notes and essays, by the late B. Jowett and Lewis 
Campbell. 3 Vols. Roy. 8vo. Clarendon Press, 42s, 


With intro- 
Post 8vo. 





































— Claudianus. 


Corpus inscriptionum latinarum. 


THE CLASSICAL REVIEW. 


Plutarch’s life of Pericles. With introduction 
eritical and explanatory notes and indices, by H : 
A. Holden. 12mo. 360 pp. Macmillan & Co. 
4s, 6d. 

Post (Edwin.) Latin at Sight. With an intro- 
duction, selections for sight reading, and selections 


for practice. 12mo, x,210pp. Cloth. Boston, 

Ginn. 90 cents. 

Sallust. Jugurthine War. Edited, with intro- 
duction, notes and vocabulary, by Edward P, 
oo 18mo. 188 pp. Macmillan & Co, 
1s. 6d. 


Sonnenschein (KE. A.) A Greek Grammar for Schools. 


Accidence and Syntax in 1 Vol. Post 8vo. Son- 
nenschein, 4s. 6d. 
Tacitus. Dialogus de Oratoribus. Edited with 


introduction and notes, by Charles E. Bennett. 
Professor of the Latin Language and Literature in 
Cornell University. ‘College Series of Latin 


Authors.’ 12mo. xxviii, 87 pp. Boston, Ginn. 
Cloth. 80 cents. Text Edition. Paper. 45 
cents. 


Taciti Dialogus de Oratoribus. Edited with prolego- 
mena, critical apparatus, exegetical and critical 


notes, bibliography and indexes, by Alf 
man, Professor of Classical Philology in U 
of Pennsylvania. a 
Boston, Ginn. Cloth. $3. 

Thucydides. Book Ill. Edited with introduction 
and notes, by C. Forster Smith, Professor of 
Greek in the University of Wisconsin. *‘ Colley 
Series of Greek Authors.’ Square 12mo. xi, 320 
pp. Boston, Ginn. Cloth. $1. 75. Text Edition. 
77 pp. 45 cents, 

Vergil, Aeneid. Book VIII. Edited with note 
word-groups and vocabulary by John Tet'ow, 
Head-Master of the Boston Girls’ High and Lat 


in 
Schools. ‘School Classics 16mo. xii 


el Guide. 
rsity 


Square 8vo. ix, 578 pp. 


; 


Series,’ 


191 pp. Boston, Ginn. Cloth. Illustrated. 50 
cents. Without vocabulary. xii, 105 pp. 40 
cents. 

Xenophon. Anabasis, Books IL.—IV. Revised 


edition by William W. Goodwin, Professor of 
Greek Literature, and John Williams White, 
Professor of Greek, in Harvard 
12mo. lii, 572 pp. Boston, Ginn. 
Haif bound. $1. 65. 


University. 
Illustrated. 


FOREIGN BOOKS. 


Aeschylus’ Agamemnon, mit Anmerkungen von R. 
Enger, 3. Aufl. neu bearbeitet von Th. Pliiss. 
Leipzig, Teubner. 8vo. viii, 187 pp. 2M. 25. 

Aesop. Sternbach (L.) Dilucidationes Aesopiae. 
8vo. 50 pp. Krakau. 1 Mk. 50. 

(Aus ‘ Dissertationes Acad. litterarum Cracoy.’) 

Anthologia latina sive poesis latinae supplementum, 

-edd. F. Buecheler et Alex. Riese. Pars II. 
Carmina latina epigraphica, conlegit F.. Buecheler. 
Fase. I. 8vo. 398 pp. Leipzig, Teubner. 4 Mk. 

Aristophanes. Lettner (G.) Bau, Wesen u. Bedeut- 
ung des sogenannten Agons in den aristophanischen 
Komodien. Lemberg. 8yvo. 19 pp. 

Ausonius (Decimus Magnus) Mosella. Lateinisch, 
mit deutscher Uebertragung von R. E. Ottmann. 


Trier, Lintz. 8vo. 88 pp. 1M. 50. 
Larth (W.) Kaiser Zeno. Basel. 8vo. 124 pp. 
Bernoulli (J. J.) Rémische Iconographie. Vol. II. 


Die Bildnisse der rémischen Kaiser und ihrer 
Angehoérigen. ILI. Von Pertinax bis Theodosius. 
(Schluss.) With 66 plates and 8 engravings. 
8vo. xii, 275 pp. Stuttgart, Union. 24 Mk. 

Blaydes (Fred. H. M.) Adversaria in tragicorum 
graecorum fragmenta. 8vo, vii, 423 pp. Halle, 
Waisenhaus. 8 Mk. 


_ Boetticher (E.) Troja im Jahre 1894. Enthiillungen 


gegeniiber dem Phantasiestiick im Deutschen 
Reichsanzeiger No. 222. Schwerin. 8vo. iil, 
34 pp. 1M. 50. ; : 
Calpurnius, Chytil (F.) Der Eklogendichter T. 
Calpurnius Siculus und seine Vorbilder. Znaim. 
8vo. 24 pp. ; 
Cicero. Dieckhoff (O.) De Ciceronis libris de natura 
deorum recensendis, Gottingen. 8vo. 79 pp. . 
—— Hildebrandt (P.) De scholiis Ciceronis Bobi- 
eusibus. Berlin. 8vo. 63 pp. ? 
— Stoerling (G.) Quaestiones Ciceronianae ad 
religionem spectantes. Jena. 8vo. 41 pp. 
Arens (E.) Quaestiones Claudianae. 
Miinster. 8vo. 42 jp. ; 
Clemens. Arnim (J. v.) De octavo Clementis Strom- 
ateorum libro. Rostock. 4to. 16 pp. 
Vol, VIL: 


Pars 


IV. Fase. I. Folio. pp. v, et 2459—3601. Berlin, 
Reimer. 58 Mk. 

Contains: Inscriptiones Urbis Romae latinae, 
collegerunt W. Henzen et J. B. de Rossi, Eng. 
Bormann, ed. Ch. Huelsen. P. 1V. Fase. I. 

Curtius. Damsté (P. H.) Lectiones Curtianae. 
Leiden. 8vo. 31 pp. 

Curtius (E.) und Kaupert (J. A.) Karten von Attika, 
1: 25000, auf Veranlassung des deutschen Archae- 
ologischen Instituts aufyenommen, Mit erliiutern- 
dem Text. Part VIII. (last). 5 maps. 46, 5x40 
em. Berlin, Reimer. 13 M. 

Demosthenes, Haar (B. ter) Demosthenis quae fertur 
oratio adversus Polyclem annotationibus illustrata. 
Sneek. 8vo. viii, 154 pp. 

Ehe (Gust.) Abriss der Kunstgeschichte des Alter- 


thums. In synchronistisch vergleichender Dar- 
stellung. Royal 8vo. xx, 675 pp. 4 plates, 557 
engravings. Diisseldorf, Schwann. 26 Mk. 


Epictetus. Zahn (T.) Der Stoiker Epiktet und sein 


Verhiltnis zum Christentum. Erlangen.  4te. 
27 pp. 
Euripides. Swoboda (A.) Beitriige zur Beurtheil 


ung des unechten Schlusses von Euripides’ Iphi- 
genie in Aulis(Fortsetzung und Schluss). Karlsbad. 
8vo. 17 pp. 

Eusebius. Heinrici (C. F. G.) 
in der Kirchenzeschichte des Eusebius, 
4to. 79 pp. 

Flensburg (N.) Zur Stammabstufung der mit 
Nasal-suffix gebildeten Prisentia im Arischen uod 
Griechischen. 8vo. 72 pp. Lund. 2 Mk. 25. 

Frank (K.) Bemerkungen zar Chronologie der Pen- 
tekontaectia. Schinberg. S8vo, 22 pp. 

Gantzer (P.)  Verfassungs- und Gesetzrevision in 
Athen vom Jahre 411 bis auf das Archontat des 
Eukleides. Halle. 8vo. 67 bp. 

Handbuch der Klassischen Altert iumswissenschaft, 
herausgegeben von Iwan von Miiller, Vol. VI. 
Part III. S8vo. xx, 625—953 pp. Mlinchen, 
Seck, 5 Mk. 50. 

Contains the conclusion of ; Sittl (K.) Archaeo- 
logie der Kunst. 


Das Urchristentum 
Leipzig. 


144 


Helbig (W.) L’épopée Homérique expliquée par les 
monuments. Traduction francaise de F. Trawinski, 
avec introduction par Max. Collignon. Paris. 8vo. 
xv, 605 pp. 

Hermann (K. Fr.) Lehrbuch der griechischen An- 
tiquitiiten, neu herausgegeben von H. Bliimner 
und W. Dittenberger. Vol. II. Part I. 8vo. 
ix, 183 pp. Freiburg, Mohr. 4 Mk. 


Contains: Lehrbuch der griechischen Rechts- 
alterthiimer, 4 Auflage, bearbeitet von Th. 
Thalhein. 


Homeros, der Blinde 


Homer. Kniotel (A. F. RB.) 


von Chios und seine Werke. Vol. I. 8vo. xiv, 
378 pp. Leipzig, Grunow. 4 Mk. 50. 


— Menrad, iiber die neuentdeckten Genfer Homer- 


fragmente und den Werth ihrer Varianten. 
Miinchen. 8vo. 18 pp. 


tibbeck (W.). Homerische Formenlehre. 3rd 





Edition. 8vo. vi, 85 pp. Berlin, Rockenstein. 
Cloth. 1 Mk. 80. 
Horatius. Gemoll (W.) Die Realien bei Horaz. 


Die Familie 


Part IV. (last). Das Sacralwesen. 
188 pp. 


Gewerbe und Kiinste. Der Staat. 8vo. 

Berlin, Gartner. 3 Mk. 20. 

Mészaros (F.)  Quaestiones de epistola ad 
Pisones Q. Hovatii Flacci. .Klausenburg.  8vo. 
56 pp. 

Tsocrates, 
auctoritate. 

Joseph (Dr. D.) 
mit Riicksicht auf 





Drerup (E.) De codicum Isocrateorum 

Leipzig. 8vo. 59 pp. 
Die Paliiste des homerischen Epos 

die Ausgrabungen Heinr. 
Schliemanns. 2nd edition. Berlin, Siemens. 
Svo. vili, 107 pp. 2 plates. 2M. 

Josephi (Flavii.) Opera, ed. apparatu critico instr. 
Ben. Niese. Vol. VI. 8vo. Berlin, Weidmann. 
26 Mk. 

IV. De bello Judaico libros VII edd. J. A. 
Destinon et Ben. Niese. 1xxvi, 628 pp. 

Jiinemann (A.) De legione Romanorum I. adiutrice. 
Leipzig. 8vo. 39 pp. 

Kern (O.) *Die Griindungsgeschichte von Magnesia 
am Maiandros. Eine neue Urkunde, erliutert. 
Royal 4to. 27 pp., 1 facsimile plate. Berlin, 
Weidmann. 4 Mk. 

Lewy (H.) Die semitischen Fremdworter im Griechi- 
schen. S8vo. Berlin, Gartner. 7 Mk. 

Limes, der obergermanisch -ratische, des Romer- 
reiches. Im Auftrag der Reichslimes-Commission 
herausgegeben von O. von Sarwey und F. Hettner. 
(In 40-50 parts.) Part I. Heidelberg, Petters. 

Engravings and 6 plates. In 


4to. 27, 13, 8 pp. 
portfolio. 5 M. each part. 

Livius. Soltau (W.) Livius’ Quellen in der dritten 
Decade. S8vo. ix, 148 pp. Berlin, Mayer & M. 
3 Mk. 


Luerctius.  Jezienicki (M.) Quaestiones Lucreti- 
anae, Aus ‘ Eos,’ 8vo, 28 pp. Lemberg. 1 Mk. 

Maass (O.) Wleitarch und Diodor. Eine Quellene 
untersuchung. I. 8vo. 83 pp. St. Petersburg. 
1 Mk. 50. 

Menander. *AmoocroAdmovAos, M., Mévavdpos TMporéx- 
twp Myintns Ayadiov. Erlangen. 8yvo. 40 pp. 
Meyer (K.) Untersuchungen zur Geschichte der 

Gracchen. Halle. 4to. 31 pp. 
Mommsen. Romische Geschichte. Vol. V. Die Pro- 
vinzen von Caesar bis Diocletian. With 10 maps 


THE CLASSICAL REVIEW. 


by H. Kiepert. 4th edition. 
Berlin, Weidmann. 9 Mk. 

Monumenta Germaniae historica. 
quissimorum tomi XIII. Pars I. 
Berlin, Weidmann. 8 Mk. 
12 Mk. 

Contains: Chronica minora saec, IV, V, VI, 
VII, ed. Th. Mommsen. Vol. III. Fase. I. 

Mueller (Lucian.) De re metrica poetarum latinorum 
praeter Plautum et Terentium libri VII. Acced. 
opuscula autoris IV, Ed. secunda. S8vo. xii, 
651 pp. St. Petersburg. 14 Mk. 

Miller (Prof. Dr. E.) Sokrates in der Volksversamm- 
lung. Zittau. 8vo. 16 pp. 75 Pf. 

Olympia. Die Ergebnisse der von dem Deutschen 
Reich veranstalteten Ausgrabung, herausgegeben 
von KE. Curtius und Fr. Adler. Vol. III. of the 
letterpress, I. half volume, and Vol. III. of the 
plates. Royal 4to. and Royal Folio. Berlin, 
Asher. 300 Mk. 

Contains: Die Bildwerke in Stein und Thon, 
bearbeitet von G. Treu. (First half: 137 pp., 1 
map, and engravings.) With volume of plates 
(69 plates in heliogravure, 1 leaf of letterpress. ) 

Peyre (R.) Lempire romain. Square 8vo. 323 pp., 

7 Fr. 50 


8vo.  vili, 659 pp. 
Auctorum aunti- 

4to. 222 pp. 
On writing paper. 


engravings. Paris, May et Motteroz. 
cents. 

Plautus. Leo (Fr.) De Plauti Vidularia commen- 
tatio, 8vo. 19 pp. Gottingen, Dieterich. 50 Pf. 


Programm (54.) zum Winckelmannsfeste der archaeo- 


logischen Gesellschaft zu Berlin. 4to. 19 pp., 5 
engravings, 2 plates). Berlin, Reimer. 2 Mk. 


Contains: Kekulé (R.) Ueber einen bisher 
Marcellus genannten Kopf in den Kgl. Museen. 
Reinach (S.) Antiquités nationales du Musée de 
Saint Germain en Laye. Bronze figures de la 
Gaule romaine. 8vo. xv, 385 pp., 1 plate, 600 

engravings. Paris, Didot. 10 Fr. 

Ribbeck (O.) Geschichte der rémischen Dichtung. 
I. Dichtung der Republik. 2nd edition. 8vo. 
ix, 358 pp. Stuttgart, Cotta. 8 Mk. 

Rochi (H.) SImagines inscriptionum graecarum anti- 
quissimarum, in usum scholarum. 4to. iii, 87 
pp. Berlin, Reimer. 6 Mk. 

Schwartz (E.) Die Konigslisten des Eratosthenes 
und Kastor mit Excursen iiber die Interpolationen 
bei Africanus und Eusebius. 4to. 96 pp. Gét- 
tingen, Dieterich. 10 Mk. 

(Aus ‘Abhandlungen der Kgl. Gesellschaft der 
Wissenschaften in Gottingen.’) 

Studien (Leipziger) zur classischen Philologie, heraus- 

gegeben von O. Ribbeck, H. Lipsius, (Wachsmuth. ) 


Vol. XVI. 8vo. iii, 171 pp. Leipzig, Hirzel. 
6 Mk. 

Toutain (J.) De Saturni dei in Africa Romana 
cultu. S8vo. 143 pp. Paris, Belin. 


Winter (Frz.) Ueber die griechische Portriitkunst. 
8vo. 26 pp., engravings. Berlin, Spemann. 
3 Mk. 

Weorner (Emil.) De Ariaetho et Agathyllo fabulae 
apud Arcades Aeneiae auctoribus. 8vo. iv, 26 
pp. Leipzig, Hinrichs. 1 Mk. 

Xenophon. Rafner(O.) Die sogenannte Socratische 
Methode, dargestellt nach Xenophon’s Memora- 
bilien und auf deren Anwendung im Gymnasial- 
unterricht gepriift. S8vo. 27 pp. St. Paul. 


The Classical Review 































d By the kind permission of the Editor I 
am allowed to put forward some further 
observations on the use of Greek Jussives, 
partly to defend, partly to explain my 
comment on Prof. Sonnenschein’s treatment 
of this subject. There are so many theories 

in vogue on this question that it will be 
advisable, at the very outset, clearly to 
define what is universally admitted as 
certain and so lies beyond discussion, and 
what still remains matter of dispute. On 
two points it may be fairly said there is 

a& consensus of opinion among. scholars. 
1, In all moods, except the indicative, 
ae. the infin.,' imper., opt., subj., the present 
and aorist have a common difference. 2. This 
common difference is not a time-distinction, 

_ as Buttmann put it, as far back as 1854 :— 
-‘Dahingegen bezeichnen die Modi des Prae- 

sens und Aorists durchaus keine Zeit.’ 

Thus as regards the identity of difference 
between present and aorist for all the so- 
_ ¢alled subordinate moods—why the impera- 

tive should be classed as subordinate one 
fails to see—as well as concerning the 


is but one view. It is only when we come 
to inquire in what this difference consists, 
_ that we find considerable variety of opinion. 
Confining our attention to the imperative 
and subjunctive jussives we will first cite 
home authorities. 
_ Mr. Sidgwick? sees the difference referred 
1 The case of indirect statement calls for special 
treatment. 
_ * Introduction to Greek Prose Composition (3rd 
edition, revised), p. 77 § 141 (1). 

NO. LXXVII. VOL, 1X. ~ 


_ negative character of this difference, there. 


APRIL 1895. 


GREEK JUSSIVES. 


to in ‘the rather fine distinction between 
the act regarded as a single occurrence, not 
considering it as protracted (aorist), and the 
act regarded as extended in time (present).’ 
Farrar has ‘ py mpaéys don’t do it—of 
momentary and single actions.’ Clyde, 
among other things, says of Aafé...xai 
dvayvwht: ‘the taking is momentary in its 
own nature and therefore expressed by the 
aorist.’ In Curtius* (Smaller, translated) 
we meet inter alia with ‘pi mparre of a 
continued action ; 2) zpagys of a momentary 
action.’ 

Rutherford :° ‘on the whole the present 
is used when the command or prohibition 
concerns an action continued or recurring, 
the aorist when it concerns a single or 
transient action.’ Prof. Sonnenschein ® has 
what Gardner Hale would call a ‘canon 
with an application,’ the latter being to the 
effect that the aorist is ‘used in commands 
applicable to a single occasion,’ and the 
present ‘in general rules of life.’ Professor 
Goodwin’s distinction is: ‘The present 
and aorist here differ only in this, that the 
present expresses an action in ite duration, 
ze. as going on or repeated, whilst the aorist 
expresses simply its occurrence, the time of 
both tenses being otherwise the same. 
And among the examples are rol roto, 


3 Greck Syntax with a Rationale of the Con- 
structions. James Clyde, M.A., LL.D. (5th edition), 

44 Smaller Grammar of the Greek Language. 
Dr. George Curtius (13th edition), p. 161 § 256. 

5 First Greek Syntax. W. Gunion Rutherford, 
M.A., LL.D., p. 95 § 229. 

6 Syntax, 341a, obs, 1. 


146 


do this (habitually), roinsov TovTo, do this.? 
Thus the aorist, according to these authors, 
would be appropriated to commands and 
prohibitions affecting :—an act regarded as 
a single occurrence, momentary and single 
actions, momentary action, single or transient 
action, and lastly occasional action. 

Now I venture to submit that, in the 
best Attic writers, the present is used just 
as freely as the aorist of actions of the kind 
just described, be they momentary, momen- 
tary and single, single occurrences, single or 
transient. Momentary is a relative term 
and can scarcely be measured by a fixed 
standard. We cannot, for instance, assign 
the decimal of a second or minute through 
which an action must have extended to 
be called continuous rather than momen- 
tary. Nevertheless there are certain actions 
which, without any strain of the imagination, 
may be regarded under the one aspect or 
the other, according to the point of view 
from which they are considered. A short 
run on a bicycle may be said to be momen- 
tary as compared with a day’s fishing, 
and the latter as compared with a three 
months’ holiday. Certain actions are more- 
over naturally regarded as continuous, 
e.g. such as involve protracted exertion, or 
such as imply a state or permanent condition. 
Thus we should naturally expect €x’ novxos, 
rapievé pot. On the other hand there are 
many actions which it is difficult to regard 
otherwise than as momentary. This is the 
case with some acts of the will and intellect, 
and even certain physical actions which are 
performed, so to speak, in no time. Thus 
the action of a warrior alighting from a 
carriage would by most persons be regarded 
as momentary. But we have Aesch. Ag. 
906 (Paley 875) &Baw’ dajvys rnooe. Simi- 
larly the act of striking, of which we read 
so often in Aristoph. wate, waite. Hach 
stroke may not have taken a second, and 
it is idle to translate ‘go on striking,’ for 
the word is found repeatedly where the 
verberation had not yet started. So of 
cessation of activity wate wade common in 
Aristoph. and éye ‘stop’—in Plato and 
Aristoph. So too the command to be off; 
dmb, dure constantly recurring, as well as 
the polite ype, xwpetre. Again the pro- 
hibition or command to approach-— 

Eur. Med. 91 

pay weAaLE pytpr Svobvpovpery. 
1b, 896 


a , / lal , , 
@ TEKVGA, TEKVO dedTE NeuTreTeE OTEYyAs. 


1 Moods and Tenses § 87. This, and Professor 
Sonnenschein’s canon will be more fully dealt with 
when we come to discuss German views. 


THE CLASSICAL REVIEW. 


Very momentary is the act indicated in 
Ar. Eq. 909 i80b 8€xov Képxov Aayd—and 
still more that of the line immediately 
following. The act of giving consent, 
marked by éa, is to my mind momentary. 
Of the two following concessions of per- 
mission which is the more continuous ? 


Ar, Pax 650 
"AXN’ Za Tov avdp’ éxeivov, ovmep €or’, elvar. 
KOTO. 


Eur. Med. 1057 


+ > \ > s a f 
gacov aitovs, ® TaAav, PEeioal TEKVOV. 


If one wishes to indulge in gratuitous 
assertions it is possible to say that in the 
following lines the present is used because 
the action of asking is anticipated to be 
somewhat protracted on account of probable 
resistance : 


Eur. Med. 940 
airod Kpéovra tivde py pevyew xOova. 
Ib. 871 


ixerever, eLarreiabe pry pevyew XOova. 


Another expression common in Aristoph. 
and Plato is 16: viv dpa, Again it is hard 
to say which is more momentary of 577 C 
(Rep. ix.) 0 89 Gd oxorer and 167 B (Charm. 
§ 32) id 8) oxéya. Somewhat similar 
is 585 C (Rep. ix.) &de de xpive. So 
Laches 190 D GAN ovtrw rordpev, & &., os 
ov Bovrda and Jb. 201 C dAAG pot odTwCt 
motnoov' aipiov ewhev adikov oixade... Where 
zowopev refers to a modus operandi previously 
suggested and zrotyaov to proceedings about 
to be suggested. 

However, if any one chooses, he may 
persist in regarding all the actions we have 
been discussing as momentary relatively 
to some infinitesimal fraction of a second. 
And, when all has been said, there remains 
a certain foundation of truth in this theory 
which will be afterwards accounted for. 
Its further development—limiting the 
command or prohibition of single, transient 
and occasional acts to aorist jussives— 
is altogether devoid of foundation, seeing | 
the almost countless instances where we find 
the present employed of such actions. 

Perhaps the best way to enable the 
reader to judge for himself in this matter 
will be to put forward parallel passages 
containing commands and prohibitions ex- 
pressed respectively by aorist and present. 
It will then be found that the present is 
used of particular, individual, definite and 
transient occasions, where the tenor of the 
order given can by no means be said to be 





ees tee 

















THE CLASSICAL REVIEW. 


general or habitual, and that on this score 
there is not a shadow of difference between 
the two tenses. 

1. wat kadar Xappidnv. Charm. 155 B. 

tht viv KaXeoov por tov Ac’. Ar, Pax 195, 
These two are absolutely particular: ‘ sum- 
mon Charmides,’ ‘summon Jupiter ’—/ic et 
nunc. In neither case is there a trace of 
habitual or repeated action. Objectively 
one is not more continuous 
other. 

2. KAA.......€kéXeve yotv viv o7 épwrav 6 T1 
tis BovAouro... 30. "H Kadds NEyers. "OQ Xaupe- 
dav, €pod airov. X. Ti epwpar; &. “Ootis 
éoriv. Gorg. 447 C. 

"AANA od, ci BovrAa, epod airov. Lb. 
448 D. 

IL. "Epéra. X. Epwrd 5. Lb. 448 B. 

"ANN ei doxet rovtoii, duadéyou Kal epwra 
6 7t BovrAa. 458 E. 

"ANN eciep BovrAcr wvdecOar, puta d7otov 
popuov... 463 C. 

Kat vdv 6 tTovrwv drorepov BovAe role 
é€pwta i) aroxpivov. 462 B. 

The foregoing commands—épod ‘ interro- 
gate,’ épwra ‘put questions’ ‘inquire ’—all 
refer to very definite and particular oc- 
casions. 

3. Quite similar is Plato’s use of dzo- 
Kptvov and azroxpwa. 

Oddéev, GAN éxretdy) od Bove, dGToKpivov 
(448 B) do you answer (hic et nunc). 

Mi OxKveu amoxpivacba, © Idde...adrja 
yevains TO Oyo, GoTep iatpd, Tapéexov 
GmwoKkpivov kal 7 pdb ) py & epwrd. 
475 E. 

“Odev otv arédumes, GT OKpivor, El ovx 
Gpa maverar dufav exactos Hav Kat 75dpevos. 
497 C. 

Cf. also 462 B supra. 

1c 8% prow azrdxpwor ovTw Kal Tept THs 

io. 8p p pl Ti 
pytopixys. 449 D. 

Taxa 51 eicopar caperrepov GAN’ droKpwau 
clot ypiv réxvar. 7 yap. 450 C. 

ob pev ov, ® &., drdxpwat TAUTO TOTO. 

Cf. Apol. 25 (where there is no differ- 
ence between dméxpwai and droxpivov if the 


latter reading is certain in 25 D).! 


The distinction between particular and 
habitual, momentary and _ repeated, is 
worthless when tested by Plato’s use of 
mépavov and zrépave. 

451 A i6c viv cal ov THv droKpiow jv 
qpopny Scam épavor. 

522 E aad’ éretirep ye Kat Tada émépavas, 
Kal TOUTO 7 épavov. 

506 C Nee, & ’yale, aitds kai méEparve 

1 Groups 2 and 3 were supplied me by the Rev. 


St. de Dunin-Borkowski, whose theory of jussives 
I hope to discuss in my next article. 


than the ~ 


Cr. Bur. Med. 701 Olowet ) aro FT 
TEepawe jeot Adyov. 

4. “Axove)(dxovgor, 

Gorg. 458 E “Axove 6y...4 Oavpalw ev ois 
Acyopevots oro cov. 

lb. 506 C “Axove on éé apxis éj00 dvada 
Povros tov Aoyov. 

Lh. 523 A "Axove by)... pdAa KaAov Aay 

Lb. 453 A “Axovcrov 7), PH lopyia. eyu 
VEPs. 

Apol, on 
Evy BeByxora. 

Ib. 20 D ’Axovere bn (alibi passim). 

Legg. B, 652 I. 5B ory Kal OTWs dKovwpe 
TPOTENOVTES TOV VOUV. 
‘ Ar. Pax 551 dxovere dew" TOUS yeopyous 
a7Tleval. 


Ov. 


Axovoare by) pou Ta €p.ot 


7 Tb. 664 dkxovtoad’ ipeis dv Evexa poppy 
EXEL. 

Lb. 670. iO vov aKOVCOV olov aprt pe Pero. 
ftem 679. Cf. Bq. 1014, 1036 

Ib. 785. jv de...p90 ixaxorvons.... 

None of the foregoing refer either to 
habitual or repeated action ; metre would 
allow of present or aorist in ll. 664, 551 
from Aristoph. Paw. 

5. Undileade)(WUndicacbe. 

Thue. 1, 86, 5 wUndileobe otv akiws rijs 
Srdprtys. 

Tb. 1, 124, 1 adda vopioavres és dvayxny 
ddixdar, & avdpes Evppaxor, Wydicarbe roy 
TOAEMOV. 

Ib. 6, 14, 1 wat ob, & zpvravt, tadra... 
erulnduce. 

6. "AyyedAe macdv dbdwrdrnvy ene 
Hee. 423. 

ravr’ dmdyyedov roca Aesch. Ag. 587 
(Paley). 

7. orovdiv AaBe 3) Kat oreicov ‘Ayabod 
Aaipovos. Ar. Lg. 106. 

dda orepavod Kal orévde TH Koadduw. 
Ib. 221. 

8.: Thue, (5, 111,96 
évOupeicbe. Id. 6, 80, 5 oxoreire oty Kal 
aipetabe 75 7)... 

oxevarbe Thuc. 1, 33, 2; 1, 143, 6; 
3, 39, 9; 3, 46, 2; $, 47, 1; 3, 58, 6; 
3, 62, 3; apud oratores passim, apud 
Plat. saepius, ef. Apol. 21 B; Ar. Pax 888 
addressed to the Povdy} and IIpurdvas. 
érioxeat, oKeyat, érurxeyuopeba, rxeyapeba 
constantly recur in Plato side by side with 
erurKoTaLEv, oKOTOMLEY, oxorupeBa, oKore, 
with practically no difference of meaning 
and all referring to matters calling for 
immediate consideration. 

9. Gorg. 482 A py Oavipate ore dym Tatra 
Adyo. ee ; 
Legg. A, 637 © was yap re i ONE 
2 , , ‘ f 4 4 
épet Oavpagovre févy..... Mi Oarvpate, © fire 


Lo 


Eur. 


- . 
OKOTELTE OVW KG@l... 


148 


Tim. 29 © éay odv...py dvvatol yryvopeba 
TavTN TAVTWS...AdyouS..aTodOdVal [7 Oavpaoys. 

Aesch. Ag. 851 (Paley) pnde Oavpacns tdde. 

10. Thue. 4, 118, 6 idvres és Aaxedaipova 
duddoKere. 

Id. 1, 86, 4 ds yuds mpérer Bovrcver Oar... 
poets didackerw. 

Ar, Eq. 202 was otv mpos ene adr’ 
eotiv; avadidacké pe. 

Gorg. 507 A. ot 8 et exes Sidacke. 

Euthyph. 6 EK ravryv roivey pe airy 4&i- 
dagov TH idéay. 

Lb. 9 A i roivwy, & dire... 
form me). 

Ar. Eq. 152 kai tov xpnopov adbrov dvadi- 
dagov as exe. 

Ar. Pax 602 rot?’ judas didagov, & Oeav 
evvovoTaTe. 

11. Gorg. 482 voule roivey Kat map’ enod 
xpHyvat eTepa ToadTa akovewv. 

Thue. 7, 77, 6 xat nv dvtAaBdpeda tov 
piAtov Xwptov...ndn vouilere ev TO exupa Elva. 

Xen. An. 6, 6, 24 vourke dvdpa dyabov azo- 
KTELVOV. 

Ar. Pax adda voile mavras dprvyas oiko- 
yeveits. (This may be regarded as general.) 

Thue. 1, 82, 4 ju) yap GAXo te vopionte. 

Id. 1, 42,1 Kai pu) vopion Sica pev rdde 
Neyer Oar... 

Id. 3, 13, 5 vopion Sé pndets... 

Id. 4, 18, 4 roy d& roXepov vopiowsor p7y.... 

Id. 7,-68, 1 épyn zpoopigopev Kat vopi- 
owe... 

Id. e 22, 1 ra dé zap’ ‘Eyeoraiov eToua 
vopicate kat A\dyw av pddtoTa €roipa evar. 

Gorg. 455 D tim éuod oy dvepwrapevos 
vopucov Kal im éxeivwv avepwrac bar. 

Cf. Apol. 18 D déuéoare....cat oijPyre and 
compare Laches 197 D pi pevtou otov pe 
apnoev oe...d\AG TpooeEye... 

Notice also that the constantly recurring 
doxetro dé pndevi is practically identical with 
mapdorty o€ pndevi, broAaBn Se pundeis, vouton 
dé pnoets, pdels oieo bw. 

12. Thue. 7, 63, 5 dpuivacbe aitods Kat 
detEare... 

Ib. 2, 89, 13 dpivacbe 8: rovtovs d€iws Tov 
Tpoeipyaopevav. 

Ib. 3, 14, 1 erapovare MourAyvators. 

Tbs Ol, 6 dpvvate obv Kal TO TOV “EAAjVov 
VOMe. 

Ar. Hq. 244. add’ apivov xaravacrpéepov 
mwaAw. 

Ib. 246 aN’ apivov Kat Siwxe Kal tpomiy 
avTOU ToLOD. 

Thue. 1, 43, 3 kal rovode pjte Evppdyors 
déxeaOe pnt’ apivere adtots dBixodow. 

13. Thue. 1, 78, 1 Bovdrcverde otv Bpadéws 
(scil. hic et nunc). 

Ib. 1, 83, 2 pnd erexOevres Bovredowper. 


.dioagoy (in- 


THE CLASSICAL REVIEW. 


Ib. 4, 87, 4 wpos ratra Bovdrcveobe cb. 

Charm, 176 © zpos ratta od ad Bovdevov. 

14, Eur. Androm. 923 wéubov pe xopas 
THOS OTOL TpoTwTATH. 

Thue. 1, 144, 2 viv d& rovrous amoKpwa- 
JLEVOL drromempopev. 

6. 1, 85, 3 xpos Tovs ‘AO. TELTETE LEV... 
TéeMTreTE O€.. 

Ib. 4, 63, 1 rodeuiovs éx tis ywpas dzo- 
nbumaueh 
Lb, 6, 34, 1 wéurmpev tpéo Bets. 

Ib. 6, 34,9 wéumwpev dé Kat és Aaxedaipova. 

Ib. 1, 128, 9 ei otv ri ce rovtTwv dpéoxe 
TéuTe avopa murTov él Oadaccay. 

15. Eur. Med. 623 ydpea. 

Ib. 1076 ywpetre, ywpetre. 

Lb. 1053 ywpetre raitdes és dopovs. 

Thue. 7, 77, 4 ywpetre, advance, 

Ar. Pax 83 pH po croBapds yoper Alav. 

Ib. 154 GAN aye Ijyace xaper xatpwv. 

Ib. 301 dedpo ras xwper tpoOiuus evOd Tis 
ocwrnpias. 

Ib. 555 adda ras xdper Tpds Epyov eis d&ypov 
TALWVLCUS. 

Thue. 4, 95, 3 ywpyoare oty ais és abtovs 
THS Te TONEWS...KaL TOV TATEpw. 

Ib..2, 72, 6 avrot de peTaXwpyoare Oot 
Botheabe, ews Gv 6 7oAEMOS 7. 

How many commands and prohibitions 
applicable to general rules of life are to be 
found in the foregoing fifteen groups? 
Not one. How many refer to habitual 
action ? Perhaps one. 

It would be idle to pursue our inquiry 
further. Suffice it to add that an examina- 
tion of similar collections for BAéze)(BAeWor, 
did0v)(d0s, €a)(€arov, ppale)(ppacov, dpuvve 
\(Gporov, dpa)(idé, AduPBave)(AGBe,  Kpive 
\(kpivat, mate)(ratoov, ote.)(roincov, tide 
)(Oés, has yielded a like result as regards the 
theory under discussion and proves beyond 
a doubt the distinction of general versus 
particular, habitual versus occasional action, 
as applied to present and aorist, to be utterly 
nugatory. 

At the risk of being tedious I must add 
another argument. The present and aorist 
often occur side by side in the same sen- 
tence, referring to the same _ particular 
occasion and sometimes even to the same 
action, ¢.g. :— 

Eur. Med. 1251 adda vu, & daos duoyevés, 
KATELIpye, KaTaTavoV, eX’ oiKkwY TAAaLaV. 

Ar. Pax 516 pi) viv advdpev add’ erevtet- 
Vopev. 

Thue. 1, 42, 
pay vopion.. 

Plato Lach. 190 pay Toivev rept oAns 
Gpets evOews oKoTdpeba...dXAa pépovs TLWOS 
Tépt TPOTOV LOwpeEv. 


id , 3 , A 
1 vedrepds tus...a€ovTw...Kat 





§ 


THE CLASSICAL REVIEW. 


I have been at the pains to make a 
calculation of the number of times the 
present is used of particular and definite 
commands—intended to be executed hic et 
nune—in a few Greek dramas, a book of 
Plato and the Speeches of Thucydides. The 
following are the results in round numbers 
always below the exact figures. The present 
then is used in the manner described in 
Ar. Hq. sixty times, in the Pax seventy 
times, Medea thirty-six times, Plato Rep. ix. 
sixteen times, Thucyd. Speeches fifty times. 

It must however be admitted that not 
many instances of general commands are 
forthcoming with aorist, although the 
formula px zoujoys Totro is not unknown in 
the sense of ‘never do this.’ This fact will 
be accounted for as we proceed. 

Lum. 526 pir’ avdpyerov 
deorotovpevov | aivéoys. 


Biov | pyre 


149 


r ie ‘ 
Pind. Pyth. 1, 179 poy) OoAwbh ys, rf Dir, 
> , a - 
evTpameAols KEepdect. 
fs Ale. Lrg. 44 Mydev aAXAo puTevoys T pOTepov 
O€evOplov ap7reAw. 
Kur. 1.7. 106 Kat pu) podds pov Tiyv Kacey 
ViTHV OTE. 
Soph. Ajax 127 sqq. 
Tolaita Tolvev eioopav, trێpKorov 
pnoev oT Elmyns altos els Oeods eros 
“J ” » ys , 
pend OyKOV apis pnoEV ; ete 
With the Editor’s permission I hope to 
discuss in some subsequent number the views 
of German scholars, notably those recently 
put forward by Koch in the latest edition 
of his Grammar and also in Neue Jahrbiicher 
99 { 
1892, page 409 sqq. 
J. Donovan. 
1 The last four examples were privately com- 
municated to me by Mr, de Dunin-Borkowski. 


NOTES ON THE TEXT OF LUCAN. 


(Continued from p. 10.) 


I co on to consider some passages in detail. 

I 254 Cimbrumque ruentem VG [furentem 
MBEU]. There is no need of references to 
prove that rwentem will stand and make 
good sense. Bentley would have introduced 
ruentem in line 68 also. Still it is not 
without doubt that I would reject furentem. 
True, we have 250 furentum and 255 furoris, 
but the frequency of such repetitions is a 
marked feature of Lucan’s style, or will be 
till emendation has run its course. 

I 320 gladii cum triste micantes BE 
[minantes MVUG]. Here E comes to the 
support of B, and micantes may very well 
be accepted, though minantes would surely 
stand, and the text can hardly be called 
certain. 

I 588 errantis in aere pinnae VUGC 
[volitantis MBE]. The sense of both words 
is the same = ‘flit about,’ see Cicero de 
orat. ii § 23. True, the word is not common 
of birds but usually figurative. Hosius 
gives an instance from Ausonius, which is 
of the latter sort. But we can hardly doubt 
that volitantis is the Pauline text, and the 
substantive is after all not avis but pinnae. 
Of course errantis will stand, but Hosius’ 
six references (all of fourfooted beasts) are 
not happy. And in Stat. Zheb. v 604 


_ errantes per capta cubilia plumae the feathers 





are not flitting about in the air, but in the 


nest, moved by the wind ; at least I can see 
no other meaning, and the passage is not 
parallel. To say ‘ volitantis schmeckt nach 
glosse’ seems to me more than can be 
assumed. May it not also be true of 
errantis 2 Curiously enough Schol, Berol. 3 
explains pennae by avium, which illustrates 
what I said above. On the whole I prefer 
to keep volitantis. See Silius viii 441. 

II 126-8 
te quoque neclectum, violatae Scaevola dextrae, 
ante ipsum penetrale deae semperque calentis 
mactavere focos ; 
dextrae VUG[C]b et in ras.M.  vestae 
B[?M']. Francken says [J/nemos. 1591) 
that the Paris MS Ashburnhamensis has 
veste after an erasure of a letter or so, and 
the correction dextrae in marg. ; while from 
A. Genthe we learn that E has vestae half 
erased and deatrae written over it by the 
first hand. It would seem then that the 
Pauline text was vestae with the variant or 
correction ‘dextrae, while the V recension 
gave dextrae. ; 

That in vestae we have an early scholion 
on deae is, says Hosius, very clear. But 
how are we going to translate the passage 
when we have agreed to accept dextrae ? 
The efforts of the scholiasts given by Weber 
are not very successful. But they all take 
neclectum with te, either absolutely = *ig- 


150 


nored,’ ‘passed over,’ or joined with dextrae. 
And I cannot see how we are to do other- 
wise. Hosius however takes violatae dextrae 
as an explanatory epithet of Seaevola = 
‘you, Scaevola of the blasted hand.’ He 
refers to Lucan’s fondness for these explan- 
atory additions, as i214 puniceus Rubicon, 
vi 674 puppim retinens echeneis, and the 
names of the snakes in book ix. But these 
all explain the very individual or class 
referred to: here the Scaevola of the day 
and his ancestor the hero of the hand story 
are different individuals separated by the 
lapse of centuries. Surely this interpreta- 
tion is too forced to be accepted, and viii 
223 aeternt martis Alanos, 245 placidi Colo- 
phona maris, Silius i 641 fatiferae iuvenem 
dextrae, all qualify the very thing referred 
to, and so do not help us, And when 
Hosius goes on to cite Sil. viii 383-4 nec 
dextra indignus avorum Scaevola as a similar 
compression, the only answer is complete 
denial, The avorwm makes just the whole 
difference, and Lucan’s [?] ‘kleine ungen- 
auigkeit’ receives from this passage also no 
support whatever. 

Hosius then takes neclectum with templum 
(sic), rendering, I presume, ‘in front of the 
temple and paying no respect to it. But 
this is a fearful straining of the passage, 
even with templum. And Lucan says ipsum 
penetrale.deae, from which it is surely even 
harder to detach neclectum. 

It is not much easier to make sense by 
taking dextrae [nom. pl.] of the Marians, as 
some of the scholia do. Nor does Kortte’s 
neglectu seem to me satisfactory, whether 
we read Vestae with it, as he and Francken 
would do, or dextrae as Hosius in bis altern- 
ative treatment of the passage. 

I prefer to keep the Pauline vestae and 
render thus ‘You too, Scaevola, whom the 
outraged Vesta did not protect, they 
butchered right in front of the sanctuary of 
the goddess.’ The murderers are the 
Marians, who outraged Vesta by their 
conduct. And so I find Bentley took it. I 
have only to add in illustration the curious 
fact that Cicero deor. nat. iii § 80, speaking 
of good men’who came to bad ends, mentions 
this case of Scaevola with special emphasis 
[ante simulacrum Vestae pontifen maximus] ; 
having just above said ($ 79) Zelamo autem 
uno versu totum locum conficit, cur di homines 
neglegant: ‘nam st curent, bene bonis sit, 
male malis, quod nunc abest.’ And this 
rendering suits well with the tone of Lucan ; 
for, however comforting the sermon [Seneca 
de prov.] of his uncle might be, the nephew 
could declare vii 444-5 mortalia nulli sunt 


THE CLASSICAL REVIEW. 


curata deo, though x 177 he pointedly says 
non neclecte ders. 

II 217-8 nec iam alveus amnem nec 
retinent ripae redditque cadavera campo. So 
VU et var. lect. m g. redduntque MBE et 
in ras. G. Doubtless we can supply the 
subject amnis to reddit and make good and 
obvious sense. But this is not what we 
want to know. We ask, why are we to 
reject the Pauline reddunt, and we are told, 
‘doch nicht die ufer werfen die leichen 
hinaus, sondern die wellen.’ But reddunt 
is not = ‘cast out’ or ‘eject.’ It is merely 
another instance of that pregnant use of 
verbs of which Lucan is so fond. It means 
redire sinit or facit, ‘ yields up the corpses 
to the plain.’ Compare i 103 Joniwm 
Aegaeo franget mare | =frangi sinet], v 605-6 
suumque in fluctus cort frangit mare, viii 
165-6 labor mazstus...proiecit fessos incertt 
pectoris aestus, and the well-known x 473 
aestimat. Itis practically the result of a 
personification. 

II 473 Luceriae [VC m] and Nuceriae 
| MBEUG] both present historical difficulties, 
as Hosius very well shows. This being so, 
why leave the Pauline text? Besides, there 
was a Nuceria in Umbria, and it is of these 
parts of Italy that Lucan is speaking. 
Luceria is in Apulia. Lucan has confused 
matters somehow, and we are by no means 
entitled to suspect the text on assumption 
of his infallibility. 

III 683-4 at faciles praebere alimenta 
carinae nune pice nunc liquida rapuere in- 
cendia cera, Here M!, that is the first hand 
of M, had carinas. Hosius tries to show 
that this may stand, by proving that incen- 
dia rapuere carinas is good Latin. This is 
not to the point. The words nune pice nunc 
liquida cera are to be taken into account. 
With their pitch and wax the ships quickly 
caught [rapuere] the fire. In the teeth of 
the MSS we need hardly force Lucan to say 
that the fire caught the ships with their [not 
its| pitch and wax, 

LV 284 paulatim fugit ira ferox. SoMU. 
cadit VBEGm. Here the authority is 
curiously divided, for BE side with V. It 
would here, as often, be interesting to know 
which of the correctors [m] of M it is that 
gives cadit. _Hosius prefers fugit, because 
cadat impetus amens comes in line 279, a 
very insufficient reason in dealing with 
Lucan. To go no further than his own 
article, on v 569 he says ‘an der wiederho- 
lung von pontus—ponti ist kein anstosz zu 
nehmen,’ And every editor of Lucan does 
and must notice the same thing. He then 
gives five references to show that cadit ira 





mee a 


THE CLASSICAL REVIEW, 


is elsewhere the regularexpression. ‘To me 


it seems that we read fugit here simply on 


the authority of M, and had better say so 
at once. Whether this is sufficient or 
not, Dr Hosius is better able to judge 
than I. 

IV 371-2 sed morbus egens iam gurgite 
plenis visceribus sibi poscit aquas. So BE 
andinras. U. clolepit MVGC. Here too 
MSS divide. MHosius thinks cepit would 
stand, and cites passages for capere = ‘ to 
contain,’ ‘be able to hold.’ How this bears 
on the point I do not see. Surely sibi 
cepit = ‘took for itself,’ to gratify the rage 
of thirst not yet conscious that it was now 
slaked. I cannot feel easy about the 
reading, for cepit can hardly have come 
from a gloss on poscit, while the reverse is 
at least possible. And, when Hosius cites 
Ovid rem. am. 535 bibe plus etiam quam quod 
praecordia poscunt, we cannot forget that, 
according to the point of view, this may 
either confirm a reading or betray the source 
of interpolation. 

IV 719 hoe solum metuens incauto ex hoste 
vidert. So Hosius. vidert is from V.. timeri 
BEUGC [and M with ¢ime in ras.|. Hosius 
argues against timeri that fear and careless- 
ness do not go together: incauto then 
requires vidert, for what Juba wants is to 
keep out of Curio’s sight. To which it may 
be answered that fear and caution go together 
very well: what Juba wants is to keep 
Curio incautus, and so he does not want 
to alarm him. In 736 Curio’s advisers 
urge him wt Libycas metuat fraudes, but he 
remains incautus as before. To me it seems 
that much depends on the significance of 
the erasure in M. If the correction be due 
to a later corrector, and if there is reason to 
think that time has taken the place of vide, 
the weight of M would tell for videri, which 
will of course stand. But, if the correction 
is from the first hand or an early corrector 
revising from the same original, the case is 
very different: and on this point Hosius’ 
collation gives no help. At present I think 
it best to retain ¢imeri with Oudendorp. 

It is more serious that ea depends on B 
only. MEVUGCb have ab. Oudendorp’s 
explanation of ex will not do. It must 
mean ‘ from,’ ‘on the part of.’ 

Hence Grotius’ incautus ab hoste tumeri 
[so with videri Burm.] is alluring, in spite of 
Hosius’ objections. But I do not venture 
to adopt it, agreeing with Hosius that i- 
cautus ought to refer to Curio, not to Juba. 

IV 600 tam defecta vigent revocato robore 
membra. So VU, renovato MBEG. True 
it is that revocato will stand, and den. i 214 


151] 


is a good reference. True also that the 
words would easily be interchanged: some 
of the blunders in the palimpsest fragments 
are of similar character. I do not however 
feel that these reasons are enough to over- 
throw the Pauline tradition. If it is good 
for anything, a pair of possibilities should 
not avail to upset it. For renovato is also 
possible, as the dictionaries show. To appeal 
to novo corpore [robore V| in 632 is perhaps 
not fair, for it proves nothing: so I lay no 
stress on it. 

V 52 famae veteres laudantur Athenae. 
So M and L (a mixed MS). fama VBEUGm 
and the palimpsest N. Hosius well remarks 
that vetus is often used with genitive in the 
Silver Age. He boldly follows M, and I 
daresay rightly. But the ablative is of 
course equally sound, cf. vi 225 informis 
facie [so MN, facies VBEUGm], ix 8 inno- 
cuos vita, etc. There must however be 
doubt here. We cannot lay great stress on 
N, for it gives pelagoque potens above (50), 
which, though approved by Detlefsen and 
half accepted by Steinhart, is not accepted 
by the editor, rightly I think. 

V 136-8 

seu sponte deorum 
Cirrha silet farique sat est arcana futuri 
carmina longaevae vobis commissa Sibyllae. 
fari is from V. fati MUGE and [ati in 
ras.|]B. Hosius, deserting the Pauline text, 
simply refers to Burman’s note. Burman’s 
main point is to get the opposition of si/et 
and fari. But it may be answered that 
(1) the notion of utterance is expressed, as 
much as it is needed, in the rest of the 
sentence, (2) it is not desirable to have the 
opposition so sharply marked as to throw 
much stress on si/et. For, as any one may 
see by reading the context, the emphatic 
part of that clause is sponte deorum. Four 
possible explanations of the silence of the 
oracle are given. 

I would keep fati and render ‘ or whether 
it be merely the will of Heaven that keeps 
Cirrha dumb, and it suffices that the aged 
Sibyl’s dark predictions of coming doom 
have been entrusted to you.’ The gods are 
supposed to take the line ‘you have your 
Sibylline Books, and that is enough. To 
me fari has decidedly the air of a deliberate 
correction, such as Hosius often detects in 
V, and the motive is probably that sug- 
gested by Burman’s note. 

V 383-4 summum dictator honorem con- 
tigit. So U E|t N}m v b. summo...honore 
VBG. summo...honori M and var. lect. g. 
Here we note that B and E for once part 
company. In much of the early capital 


152 


writing E and I could easily be confused. 
Hence no doubt the frequent confusion of 
these letters in the MSS [for M see Hosius 
praef, p. x]. It seems then that honore of 
VBG supports honort of M. Why Hosius 
edited summum...honorem I cannot guess. 
He now remarks that M is perhaps right, 
but the reading is after all ‘ ungewohnlich.’ 
His references are of little use. Itis a fact 
that he omits the one really helpful: passage 
Juv. viii 27-8 rarus civis et egregius patriae 
contingis ovantt. Nor does he point out that 
Lucan is fond of contingo witha dative [529, 
ii 707, iii 388, viii 844, ix 65, x 284]. Yet, 
if any scholar will but read the passage 
with its context in the light of the words of 
Juvenal and the usage of Lucan, I shall be 
surprised if he does not agree with Kortte 
that summo...honori is right. 

V 569-72 

zephyros intendat an austros 
incerlum est: puppim dubius ferit undique 
pontus. 

nubibus et caelo notus est ; si murmura ponte 
consulimus, cori verrent mare. 

This passage has greatly troubled editors, 
austros isin VUG. euros MBE. Hosius, 
like Bentley and Oudendorp, assumes that 
in 571-2 we have exactly the same pair of 
winds under other names. Then corus = 
zephyrus, notus = auster. It is to be noted 
that corus [NW] is not strictly = W, but 
notus and auster are both 8. ‘Thus the 
correspondence is closer where the reading 
is doubtful, which is not reassuring. The 
next words, gurgite tanto nec ratis Hesperias 
tanget nec naufragus oras, do not help us: 
for gurgite tanto is not ‘with a heavy sea 
from this quarter,’ but merely ‘with a 
heavy sea like this running.’ 

Now, is the above assumption right? 
That we find auster rendered afterwards by 
notus in such passages asii 454-60 etc. proves 
nothing. For on the other hand we find 
sephyrus and ewrus opposed in ili 549-50, 
and (cited by Hosius) Val. Flaccus i 639-40 
brings ewrus zephyrus and notus into conflict. 
See also for winds Lucan i 219, 406-7. In 
fact no other storm-scene will settle our 
reading here. When we look further into 
the present one we find the winds appear 
thus: 599 corus, 601 boreas, 603 aquilo 
{[ = boreas], 605 boreas, 606 corus, 608 eurus, 
609 notus. Here ewrus is mentioned by name, 
but not auster. Of course the storm need 
not, indeed does not, correspond exactly to 
the forecast of the fisherman. Bnt it surely 
does agree with it to some extent, and that 


not so as to justify the exclusion of eurus 


THE CLASSICAL REVIEW. 


I prefer therefore to keep the Pauline 
euros, and not to assume that 571-2 refers 
to exactly the same winds as 569-70, The 
fisherman may very well be meant to talk 
rather loosely ; for he is inclined to shirk 
the dangers of putting to sea in such weather, 
and hence prone to waste time. 

If however it should still be held that 
569-70 and 571-2 refer to the same wind- 
contest, it may be noted that with ewros we 
preserve in both cases the same relation 
between the directions of the winds. For 
in both the pairs [W and SE, NW and 8] 
we have an angle of 130° between the direc- 
tions. Thus zephyrus : eurus : : corus : 
notus. With aust7os we lose this, and have 
to be content with an incomplete identifi- 
cation. 

VI 237 nec vidit recto gladiwm mucrone 
prementem. So U. tenentem GP. tremen- 
tem MVBEN. ‘That premere is used of 
firmly grasping a sword is seen in iv 706, 
vii 562. That P and T are often confused, 
probably from the days of capital writing 
[so 261, where N has TEMTLA], is well 
known. No doubt prementem will stand. 
But surely so will the Pauline trementem. 
Of course it does not mean that the sword 
shook in Scaeva’s hand owing to his exhaus- 
tion. Scaeva’s collapse was put on to 
deceive. If Aulus had noticed the quivering 
of his sword-point [mwerone], he would have 
noticed also the suppressed rage and nervous 
excitement of the man. This was what 
made it quiver: real weakness and a flabby 
grip would not have shown itself thus. 

I think therefore that the Pauline tre- 
mentem gives the more vivid picture and is 
probably right. The variation is very old, 
for tenentem (possibly, as Hosius says, a 
gloss on prementem| appears in the Vatican 
palimpsest [P], while N has trementem. 
Compare ix 675-6 ipsa regit trepidum Pallas 
deatraque trementem Perseos aversi Cyllenida 
derigit Harpen. Here the flurried excite- 
ment of Perseus is rather different from 
that of Scaeva, but it is not exhaustion, and 
for the purposes of the present argument 
the passage is a good parallel. 

VI 681-2 quo postquam viles nec habentis 
nomina pestis contulit...... So U. et habentis 
MVBEGN. Hosius well points out that in 
this account of the witch’s caldron Lucan 
follows Ovid meé. vii 2.66 foll. very closely. 
Now Ovid (275) has Ais e¢ mille alvis post- 
quam sine nomine vebus and so forth. 
Therefore Lucan must have written ec here, 
not et. Fritzsche in his ‘quaestiones 
Lucaneae’ on the contrary holds that Lucan 
does not follow Ovid, wrongly, I think. 





THE CLASSICAL REVIEW. 153 


But in his defencé of e¢ he rightly refers to 
Lucan’s rivalry, the wish to outdo other 
authors, 

Now Ovid names a lot of ingredients in 
detail, and then winds up with ‘and a 
thousand other nameless things.’ He is at 
the end of his list. Lucan is in the middle 
of his. Ovid has nothing corresponding to 
Lucan’s viles, which surely “means ‘ ordin- 
ary. The question of sense is between 
(a) ordinary and so not worth special names, 
reading nec, and (b) ordinary and so having 
names, not the more awful ingredients, 
which are nameless, reading ef. In fact the 
passage may be closely connected with that 
in Ovid in either sense: for it is not clear 
whether the nameless things in Ovid are 
more or less horrible than those named. 

We must ask, to what does the quo in 
Lucan refer? If to the ingredients just 
mentioned 671-80, then we must render 
‘after adding to the above’...... ete. But 
if guo catches up the huc of 670, the mean- 
Ing is ‘ after adding to this [the virus /unare 
of 669] ordinary abominations,’ and the 
viles pestis are the very ingredients of 671-80 
which habent nomina, and are indeed of the 
sort that form the stock-in-trade of witches 
in allages. It is to be noted that, whereas 
Ovid speaks of the caldron, Lucan does not. 
This is strange, and may be merely one of 
the indications of haste which are often 
detected. : 

With the second interpretation of the 
passage [reading et] the following lines are 
at least quite compatible. The witch brings 
to bear a more subtle class of influences, 
which are all her own, They run 

infando saturatas carmine frondes 
et, quibus os dirum nascentibus inspuit, 
herbas 
addidit et quidquid mundo dedit ipsa venent: 
tum vox Lethaeos cunctis pollentior herbis 
excantare deos......[then the elements of the 
vow are given |. 
On the whole I do not feel that the case for 
mec is yet made out, and am inclined to 
acquiesce in the et of nearly all our MSS. 

VI 699, 700 caelwm matremque perosa 
Persephone nostrique, Hecate, pars ultima...... 
Here nostri is from the corrector of U [ul], 
but occurs in a few of the numerous MSS 
known. nostrae MVBEUG. Jlecate [varie 
scriptum] BEU. ecates [var. ser.] MVG. 

Hosius remarks on the forced interpreta- 
tions of the vulgate nostraeque Lecates. But 
after al] Persephone as the third or infernal 


form of Hecate [Servius ad Aen. vi 118] is 


no great stretch. And with nostri Hecate 
must be a pars of the witch. So Hosius 


quotes pars (maior, multa, etc.) nostri or mei 
in several well-known passages. But none 
of these are really parallel, and they do not 
help to clear up the relation here. " The tie 
between goddess and witch is surely between 
greater and less. Erichtho is rather a pars 
of Hecate. 

I think we must either keep the vulgate 
or read spes for pars as Oudendorp conjec- 
tured. Burman pointed out that the two 
words are sometimes confused in MSS, 
Here U has pers in an erasure, M pras, and 
the last letter of /ecates comes into the 
account. 

VII 363 quidquid signiferi conpressum 
limite caeli. So V. conprensum MBEUG 
[BEU add est]- Hosius refers to iii 253, 
Ov met.i48. But in the former place pre- 
meretur = ‘would have (the zodiac) over it,’ 
in the latter premuntur = ‘ are enclosed by 
the earth’ or ‘are on the earth with the 
heaven over them.’ The notion is that of 
A being below B, and the point of view is 
vertical. Here inclusion and extent are the 
notions contained, and the point of view is 
at a centre from which the surrounding 
space is surveyed. That conpressum will 
express this fairly well, I do not deny. But 
it seems to me that conprensum does it 
better. Compare ix 913 quas valli spatium 
conprendit harenas, Verg. georg. i 428, Silius 
ili 408 qua visu comprendere erat, ete. 
Therefore I would not reject the Pauline 
reading. The est of BEU is of course a 
blunder, caused by not catching the con- 
nexion of the following swmus. 

VII 421 omne tibi bellum gentes dedit omni- 
bus armis. So EUcmb. annis MBVG, 
Here Pauline authority is divided, and the 
error either way is easy. annis may come 
from 426 omnibus annis, as Hosius says. 
But he allows the repetition of nec gentihus 
ullis in i 82, 93, which is even more un- 
pleasant. I agree in preferring armis here, 
and follow him when he says ‘arma sind die 
waffenthaten fast gleich pugna oder bellum.’ 
But I. think we must then be careful to 
render omnibus like omni clade or opulentia 
in eg. Curtius iv 1 § 10, viii 10 § 20, iii 
11 § 20, ‘through doughty deeds of every 
kind.’ Whether annis might not stand in 
the sense of ‘year after year,’ I cannot yet 
feel sure. In 388 omnibus annis, if it means 
‘for evermore,’ is surely hyperbolical. The 
real meaning must be ‘for a long series of 
years,’ or as we say ‘ever so long. es, 

VII 451-2 astra Thyestae inpulit et subitis 
damnavit noctibus Argos. So V UGC. 
intulit MBE. I have already on vi 237 
noted the interchange of P and T. The 


154 


question here is, which word suits the pas- 
sage best. . For inpellere expressing the 
sudden coming on of the darkness Hosius 
compares i 235, iv 331, Sen. Phaedr. 963 
[955], and remarks that inpulerat [intu- 
lerat MVEGb] should be kept in iv 67. 
This may well be so, but in none of these 
passages is there a dative corresponding to 
Thyestae here. True, we may treat it as a 
so-called ‘ ethic’ dative, ‘he startled Thyestes 
by driving on the stars,’ that is, by hasten- 
ing the nightfall. But it seems to me that 
the dative comes much more naturally after 
intulit, which commonly takes it, either 
expressed as vi 760 inlatus mundo etc, or 
implied as i 470-1 etc. The sense ‘ brought 
the stars in upon Thyestes’ will then be 
very like vi 742-4 tibi...inmittam ruptis 
Titana cavernis, et subito feriere die, Aen. 
viii 246. On the whole therefore I think 
the Pauline intulit is here to be preferred. 
VIT 461-4 
parva tellure dirempti 
inde manum spectant: tempus, quo noscere 
possent 462 
quo sua pila cadant, aut qua sibi fata minen- 
tur 463 
facturt quae monstra forent: videre par- 


So Hosius. The important variants are 
these. 462 vultusque agnoscere quaerunt 
BEUg. 463 qua U, quam MVBEGP. 


Then 462 is omitted by G and added by g. 
And the order of 462-3 is as above in MBE, 
while in Vm and the palimpsest P it is 
reversed. In U they stand in an erasure 
as above, but it is detected that the line 
originally 462 began with a Q.. It seems 
likely then that the uncorrected U [U?] had 
the same order as VPm. We see from P 
that the order not followed by Hosius is at 
least very ancient. And Lactantius citing 
the passage gives inde manum  spectant 
tempus quo noscere possint facturi quae mon- 
stra ferant. Whether he had 463 quo sua 
in his text or not, does not appear, but he 
clearly connects 462 with 464. Kortte 
read them thus, adopting guam in 463. 
parva telluri dirempte 
463 quo sua pila cadant, aut quam sibi fata 
minentur 
462 inde manum spectant: 
noscere possent 
Sacturi quae monstra forent...... 

And I think this is the best we can do with 
this troubled passage. A few lines below 
comes the place where Hosius has on good 
grounds made his only transposition of lines 
[following U], a really fine instance of this 
proceeding, so seldom successful. And the 


tempus, quo 


THE CLASSICAL REVIEW. 


text of book vii is perhaps more disturbed 
than any other part of the poem. 

I would render the sense thus. ‘ Parted 
by but a little space of ground, they looked 
to see who was to be the target for their 
javelins, whose hand on yonder side [inde] 
destiny threatened to turn against them : 
they had time to learn what abominations 
they were about to commit.’ 

For tempus [ fuit] see Hosius’ references. 
For inde see 502-3 frigidus inde stat 
gladius. 

VIT 594 humanum columen, quo cuncta 
premuntur. So VGm [?U!]. reguntur 
MBE et in ras. U. Hosius gives quotations 
(all with fastigium| to support hwmanum 
columen, and adds ‘ premuntur dann ist kraft- 
iger und ausdrucksvoller als reguntur, 
obwohl auch bei Manilius i 27 quo cuncla 
reguntur.’ But the question is rather what 
Lucan wants to express. If ‘the height of 
human eminence, dwarfing all things else,’ 
then premuntur is all that Hosius says. If 
‘by which all things are gauged’ = in 
which all find their common standard, then 
reguntur is better. Neither word sounds to 
me like a gloss on the other. Yet Schol. 
Berol. 3 interprets premuntur by reguntur. 
But I take it he means to interpret ‘are 
held down’ by ‘are governed.’ In any case 
the difference is one of subtle shades of 
meaning, and I cannot feel satisfied that 
the Pauline tradition is wrong. The vari- 
ation is probably a very old one, and one 
must admit that Paulus may have chosen 
between the variants and not chosen rightly. 
It is with grave doubt that I accept pre- 
muntur. 

VIII 567 qui vetet externas terris advertere 
classes. So UE. adpellere MVB. expellere G. 
To prove, as Hosius does, that advertere will 
stand, is nothing. Who doubts it? That 
appulerat comes just above in 563 is also 
nothing in Lucan. If we do not accept 
adpellere here, it seems to me that we give 
up the case for M altogether. The vari- 
ation is probably very old. 

IX 385-6 
durum iter ad leges, patriaeque ruentis amore 
per mediam Libyen veniant...... 

Here amore comes from C.  amorem 
MVBEUG. Now the lemma of C is hardly 
first-rate evidence: the scholion is durum 
est iter quod ad leges patriae tendit et amore 
libertatis occumbit. The verbs tendit and 
occumbit are clearly parallel ; ocewmbit does 
not give the sense of what follows, which 
would rather be in magna pericula ducit or 
words to that effect. It seems to me that, 
like tendit, it is really an exposition of the 


THE CLASSICAL REVIEW. 155 


force of ad. That is, the scholion refers to 
amorem, not amore, and either the latter is 
a slip of the pen in the lemma or lemma and 
scholion (as happens elsewhere) do not agree. 
When we turn to Weber we find Schol. 
Lips.! quae omnia non debent vos dehortari ; 
nam durum est iter quod ducit ad stalum 
leyum retinendum, et ad amorem patriae 
ruentis, i[d est] ut defendatur, ne ruat. 
Whatever be the authority of this note I 
believe it is right in connecting the line 
with what precedes, not with what follows. 
C and some other scholia rightly point out 
that durum iter etc. is opposed to 393-4 
vadat ad dominum meliore via. But this 
surely assumes that the former closes the 
first member of the antithesis and the latter 
the second. In fact to read amore and 
connect with what follows is a wilful per- 
version to which MSS and scholia lend no 
support. ; 

Therefore keep amorem and explain ad 
leges [tutandas| ad amorem { praestandum). 
Cato says ‘We are entering on a course of 
hardship and peril; but if you want to 
uphold the constitution, if you mean to 
show that you love your country by trying 
to stay her fall, your road must necessarily 
bea hard one. If you shrink from peril 
and hardship, you are not the men for me: 
choose the easier path, but do not forget 
that it leads to slavery.’ 

That Lucan is fond of this pregnant use 
of prepositions is plain enough. Cf. 733 
nec vobis opus est ad noxia fata [ peragenda] 
veneno, Vv 368-9 ne tela sibi dextraeque negen- 
tur ad scelus hoe [ faciendum), x 384, vi 
616-7 aditus...patebunt ad verum [noscen- 
dum], and many more. In v 476 miscenda 
is expressed. Soi 336 post Cilicas {domitos], 
vi 145 ante feras Rhodani gentes |domitas|, 
ix 243 post te [mortuwm], and many more. 
In vii 15 domitas is expressed. 

Why does Lucan say ad leges, and not 
[ad] patriam? Because ad patriam would 
be liable to misunderstanding, ad patriae 
amorem is much clearer. For the views of 
Cato on these matters generally see ii 286—- 
319. 

IX 406-7 sic ille paventes incendit virtute 
animos. Here paventes is doubtfully traced 
to U [ut vid U] and is given as a variant 
by the corrector of V [v]. calentes MV BEGu. 
Hosius well argues against the latter. But 
this does not establish paventes. On the 
contrary, I suspect that we are not here 
dealing with two old variants, but with text 


and gloss. I think paventes is the gloss on 


the word’ hidden in calentes, which is pro- 
bably cadentes. Lucan has cadere 11 729 


(robur), iv 279 (impetus), perhaps 284 (ira, 
see above ad loc.). But Aen. iii 260 cecidere 
animi, Ovid met. vii 347, amply prove the 
usage. For the confusion of metaphor in 
Lucan cf, 1 159, 262-4, vi 7-8, ete. 

IX 454 Aeoliam rabiem totis exercet habenis. 
So G. harenis MVBEU. I will admit that 
totis habenis may mean ‘ unchecked,’ though 
Hosius’ references are inadequate, the one 
not having fotis and the other being literal 
(of a horse), not figurative. But who doubts 
it? The question is, why do we want to 
get rid of harenis? Surely not because 
lines 437 and 441 above and 465 and 469 
below end with ‘Aarenis. And in sense it 
may well be held to have the advantage : 
for totis habenis adds little to liber meatu 
preceding, while darenis is in contrast to 
the rocks and trees on which Lucan has 
just said that the wind does not spend its 
fury. See also below 466-71. Auster 
‘vents his fury on the whole sandy tract,’ 
that is, ‘all over the sand,’ on the stretch of 
open sands. So vii 506 toto diduait cornua 
campo, ‘all over the plain,’ and often. In 
short I cannot see any good reason for 
deserting the MSS here, Aarenis being in my 
view an emphatic word. I should add that 
the codex Ashb. [see on ii 126-8} also gives 
arenis, Which Francken accepts. 

IX 588 monstrat tolerare vapores. So VU. 
labores MBEG. That either word will stand 
is clear. Hosius says that iv 305 supports 
vapores, while labores comes from ix 88l. 
But the Jabores of 881 are part of the toils 
of the same march, and in iv 305 siccos is 
added to vapores. As to the present con- 
text, it is true that 591-3 refer to thirst. 
But 589-90 refer to bodily exertion. This 
seems to be a case of very old variants 
between which the MSS authority must 
decide. If not, what becomes of the superi- 
ority of the Paulines | 

X 136 discubuere toris reges. So MEU, 
illic VBG. Hosius cites several passages 
for discubuere toris, which is well enough. 
But he does not account for illic. The 
variation is surely a very old one. Did 
Lucan leave two versions, or is one of them 
an ancient gloss? If the former theory be 
accepted, then where Lucan hesitated 80 
If the latter, how can tlie be a 


may we. he | 
gloss on toris? But foris is a natural gloss 
on illic. Therefore I doubt the wisdom of 
displacing illic, which Oudendorp and Kortte 
preferred. 

X 244-5 


vel quod aquas totiens rumpentis littora Nili 
adsiduo feriunt coguntque resistere flatu. — 
Heinsius and Diels, most editors, Hosius 


156 


included, read flatu. This appears as a 
variant given by m vy, the corrector in each 
case erasing the last letter of fluctus, so that 
it is alternative to fluctu. The scholion of 
C is also given as authority for flatu. But 
the lemma gives the whole line with jluctu, 
and the scholion is ‘7d est wenti.’ Surely this 
refers to the whole line, and venti is nom. 
plural. But this is quite consistent, with 
fluctu. The wind may operate either by 
blowing against the stream of Nile or by 
driving thesea-waves againstit. Thus v 605-6 
he says of Boreas swumque in fluctus cort 
frangit mare. The authority of C therefore 
is not for flatu. And the context is for 
fluctu. There are two ways in which the 
etesian winds cause the Nile to rise: by 
bringing the rain-clouds over its catchment- 
area | fluvio cogunt incumbere nimbos|, and by 
piling up a water-barrier against its mouths. 
Hence 246-7 ile mora cursus adversique 
obice ponti aestuat in campos. In both cases 
the winds operate through the agency of 
something set in motion by them. 

I have now to remark that the MSS 
authority stands thus. flatw var. lect. m v. 
fluctuGCmy. fluctus MVBEU. It seems 


THE CLASSICAL REVIEW. 


to me that, adsiduo being taken for an 
adverb, fluctu was made fluctus. But this 
reading is after all not impossible, and the 
blunder, if a blunder, may have come about 
in another way. However I am strongly 
for fluctu. 

It is interesting to note that the editors 
who read flatu thereby make another in- 
stance of the near recurrence of words in 
Lucan, for line 240 ends with flatus. Of 
Hosius’ two references Mela i 9 § 53 proves 
that flatu, Plin. epp. iv 30§ 8 that either 
reading, will stand. 

X 536-7 dux Latius tota subitus formidine 
belli cingitur. So V. subitto MBEUG. 
Hosius, after Oud. and others, cites passages 
where matutinus [Aen. viii 465], nocturnus 
etc. are used. But the verb in these refer- 
ences is active. Here cingitur is not = se 
cingit, and such references are irrelevant. 
To me subito seems clearly right, and why 
Grotius and others should prefer swbiti is a 
mystery. Lucan is fond of adverbs, and 
several have been restored to him by Dr 
Hosius himself, 

W. E. HEITLAND. 


NOTES ON LUCRET. III. 962, AND VARRO, SAT. MENIPP. (Lumen. 16, 17). 


Lucretius De Rerum Nat. iii. 962 (960) : 


aequo animoque agedum magnus concede : 
necessest. 


This is the reading given by Munro, 
magnus in place of magnis being ascribed to 
‘Censor Orellii Ienensis.’ Lachmann adopts 
dignis, Bernays gnatis. These last are 
possibly appropriate guesses: Munro’s 
reading seems at first to be more, from its 
closeness to the MSS.; but is it really 
appropriate in sense? I doubt it: no 
cessation from tears can avail to make the 
‘balatro’ ‘ dignified’ in the eyes of Nature. 
On the contrary, a term of abuse is wanted 
here. Just as the ‘inclamet..et uoce increpet 
acri’ of v. 954 is re-echoed in the ‘ increpet 
inciletque’ of v. 963, so the imperative 
addressed to ‘ balatro’ in 955 should have 
for its echo in 962 an imperative emphasized 
in like manner. Hence for magnis read 


Maccus—the name of an Atellane character 
fitly applied to an old fool as a variant for 
‘balatro,’ which again is akin to the Atellane 
‘pucco’ (ep. Hor. Sat. i. 2, 2). 


Varro Sat. Menipp. (Eumen. 16-17): 


Ajax tum credit ferro se caedere Ulixem 
Cum bacchans si/wam incedit porcosque tru- 
cidat. 


Siluam Cod.: suile Riese: suillam caedtt 
Ellis, which, as Wordsworth remarks, is 
‘very ingenious but too like “‘porcos 
trucidat”’ to be likely’: cum baccas ferula 
caedit Biicheler. 1 think, with Biicheler, 
there is meant to be a double antithesis, of 
both object and instrument, and suggest 
silicem inlidit; ep. porcum saxo silice per- 
cussit Liv. i. 24; and Cic. Div. 2, 41, 85. 


R. G. Bury. 


oa 


4 
3 
x 





THE CLASSICAL REVIEW, 


HILBERG ON THE OVIDIAN PENTAMETER. 


Hilberg's Gesetze der Wortstellung im Penta- 
meter des Ovid. Teubner. 1894. M. 28. 


Tuis book is a natural, if in some ways a 
perverted, outcome of the many-sided study 
of Ovid which has marked the last forty 
years. Since the appearance of Merkel’s 
Tristia and Ibis in 1837 hardly a corner of 
the vast Ovidian domain has remained un- 
explored. Both the Zvistia and Jbis have 
been set on a new footing by the editions of 
8. G. Owen (1889), and my own (1881), 
each of them not only read but carefully and 
minutely studied by Hilberg: the Pontic 
Epistles were published by Korn with a full 
app. crit. of MSS. in 1869 ; the Metamm. with 
a complete collation of the Marcianus also by 
Korn in 1880; subsequently Zingerle, 
Riese, who has given a complete collation 
of the Neapolitanus, and Magnus, have again 
edited them ; Magnus indeed by various pro- 
grammes, anda continuous series of thorough- 
going diatribes in Fleckeisen’s Jahrbiicher, 
has perhaps done more than any other critic 
to adjust the respective claims of the 
MSS., a task of no little difficulty in the 
case of the Metamorphoses. Even the 
Halieutica fragment has obtained an ad- 
mirable editor in Birt, whose monograph will 
remain unsurpassed in the history of Ovidian 
criticism, though on a scale not comparable 
with the same editor’s recently published 
Claudian (1892); Palmer and Sedlmayer, 
Palmer especially, by his often jealously 
ignored but indisputably admirable emenda- 
tions, have revolutionized the criticism of 
the Heroides: Ehwald in his 1888 reprint 
of Merkel’s Amores, Ars Amatoria, Remedia 
Amoris, and Medicamina Faciei gave a con- 
spectus of the most important readings of 
the earliest MSS. of those works, of which ! 
however we still require a minuter collation, 
the more that they are indubitably the finest 
flower of the poet’s genius: on the Fasti 
Peter, Polle, and others have exhausted an 
erudition which too often fails to clear up 
the problems in dispute, as indeed most 
questions of Roman topography still remain 
debatable, or at least not determinately 
settled. 

Of all these works, so far as they touch 


_ the pentameter, Hilberg has availed himself, 


and his readers will find in him—for the 
work contains little short of 900 pages— 


1 Except the Med. Fac. exhaustively edited by 
Kunz in 1881, 


very detailed discussions of a large number 
of lines in which the reading is doubtful, 
and into which conflicting considerations 
enter. This indeed is not the primary 
purpose of the book, but it is perhaps the 
most interesting, and at any rate has the 
advantage of calling the reader’s attention 
to some of the most crucial questions, so far 
as these occur in pentameters. 

The primary object of Hilberg’s work 
is to elicit the laws which guided the poet 
in arranging the words in his pentameters. 
Any one at all versed in the Elegiac poems 
of Ovid, and a fortiori such as have practised 
the composition of Latin elegiaes (in these 
days mainly Englishmen), arrive, after a 
very short study, at the conviction that the 
rules by which the Ovidian pentameter is 
regulated are of the strictest, most rigid 
kind. That elision is of the utmost rarity 
—that the first half of the line begins 
preferably with a dactyl—that if it begins 
with a spondee, the spondee is not one com- 
plete word, but part of a word which is 
continued into the second foot—that the 
last syllable of the first half is preferably a 
naturally long syllable—and this syllable 
rhyming with the last syllable of the second 
half—so much becomes, to a really careful 
student, clear after studying the poems, as 
exhibited in most of the post-Heinsian 
editions. He sees that those Greek licences, 
e.g. allowing a word of three four or five 
syllables to end the line, and admitting the 
last syllable to be indifferently long or short, 
which Catullus Tibullus and even Pro- 
pertius still permitted themselves, are 
studiously avoided by Ovid, indeed are 
almost entirely relegated from his most 
finished poems. When he takes up the 
Pontic Fpistles, the relaxation of these 
rules accompanies an obvious decline in the 
poet’s powers and genius. I should suppose 
that most very close students of Latin 
metric—especially if they have studied 
Lucian Miiller’s de re metrica, the newly 
appeared second edition of which book I here- 


‘with commend to my readers’ notice—will 


have formulated for themselves the above, 
or most of the above, rules as distinguishing 
Ovid’s pentameters from those of his pre- 
decessors. The question is whether it is 
possible to go farther; whether we can 
formulate these and perhaps other rules with 
sufficient precision to be able to pronounce 
by an appeal to them which of several pos- 


158 


sibilities presented by the MSS. must be 
right. This is the point raised by Hilberg, 
and no one who has not examined his book 
ought to pronounce in a hurry on the 
question. 

He lays down eleven laws :— 

A. The position of the word must not 
violate the prosodial and metrical laws of 
Ovid. 

B. The more or less emphasized words 
should, if possible, be represented by their 
position in the pentameter. 

O. The natural order of the words is 
observed so far as A and & permit. It is 
only within the most rigidly defined limits 
that this order is broken through in favour 
of law H (that the pentameter should begin 
with a dactyl). 

D. An adjective stands defore the sub- 
stantive or pronoun with which it agrees, 
so far as this is consistent with 4 BC HJ. 

E. Short vowels at the end of the penta- 
meter are avoided. 

F. Appended st (est) is preferably found 
at the end of the pentameter. 

G. The first half of the pentameter ends 
preferably with a syllable long by nature, 
not by position. 

H. The first foot is, if possible, a dactyl. 

J. If the first foot is a spondee, it should 
not form a complete word, 

K. Adjective and substantive agreeing 
with it are, if possible, in different halves 
of the pentameter. 

L. The verb is placed as early in the 
pentameter as is consistent with the other 
laws. 

The first criticism which I would offer on 
this, is—that it is scarcely a right use of 
the word ‘law’ (Gesetz). The exceptions to 
most of the rules above drawn out are very 
numerous indeed, and it is just in this that 
Hilberg fails to carry conviction. Take /. 
He first obtains from a large number of 
examples proof of the tendency of Ovid to 
place sé (est) at the end of pentameters,! 
especially after a short 4 or ¢: then pro- 
ceeds to insert it in places where it has no 
support from the MSS. e.g. Vast. ii. 719, 
720 
Ile iacens pronus matri dedit oscula terrae. 

Creditur offenso procubuisse pede. 

1 And of hexameters. Thus 4. A. i. 655, 6 
Iustus uterque fuit : neque enim lex aequior ullast 
Quam necis artificis arte perire sua : where Hilberg 
remarks ‘ hier ist es Merkel, welcher das lingst ein- 
cesarkte ’st zu neuem Leben erweckt,’ and adds 
that, however, such appended st is rare at the end 
of hexameters, the syllable being im thesi, as 
opposed to the pentameter, the last syllable of which 
18 UN Grst. 


THE CLASSICAL REVIEW. 


Merkel in his later editions altered this 
with the best MSS. to Creditus. Hilberg 
goes a step farther and adds ’st after pede. 
I do not think this is right. There are 
other instances of the participle creditus 
used in this somewhat rare manner, 7.e. 
as=in quo creditus est ‘and was believed in 
doing so to have stumbled.and fallen.’ Nor 
can I believe that in Jbis 458 Victor ut est 
celeri uictaque uersa pede the poet really 
wrote Victor uti celeri uictaque uersa 
pedest, by which entirely unsupported 
alteration the verse is brought into har- 
mony not only with /, but also with C and £. 

In the well-known passage of the Aemed. 
Amoris 476, where Agamemnon describes 
Briseis as only differing from Chryseis by 
a syllable— 


‘Est,’ ait Atrides, ‘illius proxima forma, 
Et si prima sinat syllaba, nomen idem ’— 


two MSS. give idem est. Can any one say that 
idem is right, dem est wrong? or reversely ! 
Surely nothing is gained by Hilberg’s 
‘laws’ for determining such a question. 
All we can safely do, is to observe Ovid's 
general use, primarily in the A. A. and 
Rem., then in the Amores and Heroides : 
subordinately, and with a great deal of 
reservation, in his later and much inferior 
works. For I must come to what most 
lovers of the poet will, I think, agree with 
me in enforcing, and there is hardly any 
point which I would so earnestly press upon 
the attention of the ingenious writer I am 
reviewing as this—that the works of Ovid's 
prime ought to be judged by a dzfferent 
standard from those of his decline —especi- 
ally from the Zristia and Pontic Epistles.” 
Licences or irregularities which we have 
reason to think were absolutely unknown 
in the finer works of his adolescence and 
early manhood were very likely to find 
admittance in the comparatively uninspired 
elegies of his later years when the misery of 
his sudden and unexplained banishment 
combined with the dismal surroundings of 
Tomi to depress his spirit and freeze the 
genial current of his Muse. But Hilberg 
not only equalizes all the Elegiac poems as 
teaching us the ‘laws’ of Ovid’s penta- 
meter, but admits works of very doubtful 


2 The Jbis stands on a different footing. The 
metre is throughout singularly careful, and if the 
work is genuine, is quite worthy of the poet’s best 
days. Few however can doubt that it has been, to 
say the least, interpolated. I may refer sceptical 
readers to my article in, the Cambridge Journal of 
Philology for 1885, pp. 98—106. , 


THE CLASSICAL 


genuineness, such as the Nua, the later 
Heroides, and the Consolatio ad Liuiam, to 
rank on a par with the genuine. So un- 
critical a course is the more surprising, as 
the task of eliciting ‘laws’ is easier and 
simpler if these doubtful works are kept 
aloof. The Consolatio, in particular, bears 
on its face the sigas of extraneous authorship, 
and the numerous discussions which its 
peculiarities have provoked would better 
have been spared and must be considered 
intrusions in Hilberg’s volume. 

To return to /. The following passages 
seem to me to have drawn from Hilberg a 
wrong conclusion. 


Her. xii. 73, 4 
Ius tibi et arbitrium nostrae fortuna salutis 
Tradidit, inque tuast uitaque morsque 
manu. 


‘In some MSS.’ says our critic, ‘the ’s¢ 
after fwa is wanting, and it should be un- 
hesitatingly removed, for where ’s¢ does not 
obviate hiatus or lengthen a vowel (pedest 
ete.) Ovid only admits it in the inner half 
of the pentameter in those very rare cases 
where a misunderstanding would arise 
without it.’ To this I must demur. The 
*st seems to me absolutely required. Let us 
take another case. 


Am. i. 7, 20 
Ipsa nihil: pavidost lingua retenta metu. 


Hilberg would remove ’s¢ for the above 
reasons and ‘because it should naturally 


follow, not precede, retenta, as laid down in 


‘law’ C. Here we see the arbitrariness of 
the procedure. Because in a large number 
of instances est (st) follows the participle to 
which it belongs, a ‘law’ results that it 
should always do so, except where some 
palpable reason exists for neglecting it. 
‘Then the cases where the ‘law’ is violated 
are pronounced to be wrong. But what is 
the in every way more probable decision ? 
We notice (1) that the verb to ipsa is 
omitted: here is a reason why in the sequent 
clause it should not be. (2) est (st) is 
avoided after -w, therefore it is unlikely to 
have been placed after metu at the end of 
the line. If, then, it was inserted at all, it 
can only be after pavido. So much « 
priori: then, what is the evidence of MSS. ! 
Hilberg states nothing on this point; and 
I am not sure even that the readings of the 
two earliest, the Puteaneus and the San- 
gallensis, are known: for we still desiderate 
a thorough collation of MSS. in this, the 


REVIEW. 


most exquisite of all Ovid's poems, a : 
in the Ars Amatoria. Strange that the 
two works which made the poet famous in 
every province of the Roman Empire should 
still lack a completely adequate edition 
while the //eroides, in every way an inferior 
work, should have been thought worthy of 
a collation as exhaustive as the works of 
Horace, [ venture to hope that this task 
may still be undertaken by some one of the 
increasing band of palaeographers whom 
Oxford and Cambridge are training, 


Trist. iii. 10, 9, 10 
At cum tristis hiemps squalentia protulit 
ora, 
La hl am ° 
Terraque marmoreost candida facta gelu. 


It might be expected that Hilberg would 
follow the line of reasoning adopted in Am. 
i. 7,20 and omit’st. Notso: itis required, 
he says, to prevent a misapprehension. 
Without ’st, it might be thought that both 
hiemps and terra were subjects to protu/it. 
To the conclusion, that ’s¢ is indispensable, 
most readers will assent: but the ground 
alleged is inadequate: nor is it much helped 
by the wish to avoid ’st after gelu. For 
Hilberg himself admits that in two cases, 
A, A. i. 552— 


Terque fugam petiit: terque retenta metu 


est. 
Horruit, ut sterilis agitat quas uentus 
aristae— 


and Pont. iv. 1, 14 manu'st, Ovid has 
permitted this collocation. One is tempted 
to ask, are there no others! Yet if there 
are only two, the law is broken: and, so 
far as MS. evidence goes, it is quite 
doubtful whether in 7rist. iii. 10, 10 the poet 
adds ’st after marmoreo or after gelu. And 
do not let us forget that the ‘laws’ depend 
ultimately on MSS.: and that these vary 
constantly and cannot be said to give any 
certain sound, at least on this point of ‘st. 

Coming to another line discussed on p. 
413, Zrist. i. 6, 6— 


Si quid adhue ego sum, muneris omne tui est 


(which line is part of the page facsimiled 
from the best MS. of the Tristia, the 
Laurentianus (Z), in 8, G. Owen's edition), 
it appears to me very questionable whether 
st (est) should be omitted on the showing of 
the facsimile. It is true that es¢ 1s not im 
the line as written at first: but omne (uit 
is unmistakably written over, and Hilberg 


160 THE CLASSICAL REVIEW. 


speaks unadvisedly when he says that there 
is a doubt as to the meaning of +. It can 
mean nothing but est, as Owen, of course 
rightly, explains. 

An interesting rule which is formulated 
on p. 414 deserves close consideration. It 
is in reference to final 7 (in cases where the 
quantity is indifferently long or short, mht 
fibi sib’). Hilberg lays down this rule: 
Ovid omits ’s¢ (es) where final ¢ rhymes with 
iin the first half of the pentameter, e.g. 


Rem. Am. 228 
Aeger, et oranti mensa negata mihi 
Trist. ii. 104 
Cur inprudenti cognita culpa mihi? 
Trist. ii. 208 
Alterius facti culpa silenda mihi 
but Rem. Am. 582 
Est opus: auxilio turba futura tibist 
Fasti i. 480 
Siste precor lacrimas! ista ferenda tibist 
Fasti iv. 456 
Nec mora ‘me miseram ! filia,’ dixit, ‘ ubist ?’ 
Pont. ii. 9, 72 
Et tamen his gravior noxa fatenda mihist 
Pont. ii. 10, 10 
Vel mea quod coniunx non aliena tibist. 


A similar rule is* enunciated for final 
4 on p. 416: that is to say, where such 4 
rhymes, sé(est) is omitted. But here the 
case seems more doubtful: at any rate the 
MSS. exhibit very great fluctuation, Mean- 
while I need not say how greatly such 
questions affect palaeographical research. 
When a MS. of first-rate importance, like 
the Trinity College (Cambridge) codex () 
of the Jbis, or Owen’s LZ of the Tristia, 
comes to light—and of the immense weight 
of both codices Hilberg’s pages afford the 
most abundant evidence—its readings not 
only do not stand on a parallel with 
ordinary MSS., but rank among the ultimate 
standards by which such points as Hilberg 
raises have to be judged. Conversely, one 
of the subordinate yet real gains from so 
thorough a book as his is lies in the 
clearness with which it exhibits the intimate 
connexion of palaeography with almost every 
point of philological research—orthography, 
metric, grammar, mythology, archaeology. 

It is time to turn to another section of 
our critic's book. We may take law C, 
which enforces that the natural order of the 
words in the pentameter is kept so far as 
is consistent with other laws, notably // (that 
the pentameter begins with a dactyl). Here 
again it seems to me that the works of the 
poet do not all stand on the same level. 


In the Amores and Ars Amatoria the order 
of the words in the pentameter is, speaking 
generally, as nearly the natural order as 
metrical considerations permit. In _ the 
Heroides this is not so: the pentameters 
are more complex in construction, and the 
arrangement of the words less direct and 
simple. This is, as might be expected, 
equally, perhaps more, true of the T'ristia 
and Pontic Epistles. On the other hand the 
Fasti show a return to the plainer and more 
direct order of the Amores. I shall attempt 
to prove this by examples. 


The following are taken from the Amores : 


Damnabitque oculos et sibi uerba dabit 
Non caret effectu quod ualuere duo 

Centum sunt causae cur ego semper amem 
Siue rudis, placita es simplicitate tua 

Haec melior specie corporis, illa sapit 

Me miserum ! quare tam bona causa meast ? 
Maesta erat in uultu: maesta decenter erat 
Tu tamen ante alios, turtur amice, dole 
Apta quidem dominae, sed magis apta mihi 
Aerati postes, ferrea turris erat 

Liber et Alcides et modo Caesar habent 
Egressum tectis, pulcher Iule, tuis 

Perdere: non ego sum stultus, ut ante fui. 


The following from the Ars Amatoria : 


A pereant, per quos munera crimen habent 
Vir mala dissimulat, tectius illa cupit 
Quamuis sit mendax, Creta negare potest 
At puto non poteras ipsa referre uicem 
Ut fragilis glacies interit ira mora 
Perprime temptatam nec nisi uictor abi 
Oscula deinde dabit, deinde rogabit emas 
Et modo festines et modo lentus eas 

Et siquis male uir quaerit habere uirum 
Fac tantum cupias: sponte disertus eris 
Depexaeque iubae plausaque-colla iuuant 
Ei mihi! rusticitas, non pudor ille fuit 
Casus inest illis, hoc erit artis opus 

Sit tua cura sequi, me duce tutus eris 
Altius egit iter deseruitque patrem 

Nil opus est illi qui dabit arte mea 
Perfer et obdura! postmodo mitis erit 
Pummodo sit diues, barbarus ipse placet. 





THE CLASSICAL REVIEW. 161 


In all these the words follow an order 
almost identical with the order of prose. I 
suppose that it will be a surprise to many 
who have never looked into the matter to 
find that Wordsworth’s theory, that the 
natural order of words in poetry is the 
right one, is confirmed by the two most 
highly finished poems of Ovid. It is difficult 
to see why this should not be equally 
true of the Heroides, where the subject is 
the same, love. But the difference, which 
makes itself felt on the shortest comparison, 
is marked and unmistakable. The reason, 
I imagine, lies in the different way in which 
the same passion is presented. The Greek 
heroines who speak in the Heroides plead 
their cause (generally, not to be forsaken 
by their lovers) with all the arguments 
which feminine rhetoric can urge. It is no 
wonder that their language should at times 
become artificial like their reasoning, or that 
rhetoric should employ its usual methods 
of antithesis, inversion, and the other arts 
by which the diverse phases of passion find 
their habitual expression. Whereas in the 
Amores we have short idyllic scenes or 
phases of a lover’s life. It is the poet- 
lover showing us his own feelings as directly 
and plainly as he can: a male and highly 
sensuous nature expressing his not too 
refined or ideal emotions in words which 
convey their meaning without reservation 
or ambiguity. In the A. A. this is even 
more decidedly true. In this Manual of 
Love for the use of Men and Women, Ovid 
never beats about the bush; his precepts 
are straightforward and delivered in the 
most straightforward words : take away the 
single element of obscurity, the mytho- 
logical allusions, which strike a modern 
reader so grotesquely, and the work is 
intelligible to the least cultivated under- 
standing. When we come to the 7’ristia 
the case is very much altered. Here alone 
such involutions of clause, as 


Si quis, qui, quid agam, forte requiret, 
erit. 


such poor antitheta as 


Inque suis amat hunc Caesar, in hoste 
probat, 


or such inversions as 
Quam tribuit terris, pacis an ista notast? 
become, if not frequent, at least not 


uncommon. 
NO. LXXVII. VOL IX. 


This difference of SU) le in different works 
is not recognized by Hilberg, and, so far as 
it goes, seems to diminish the weight of his 
conclusions. But the reader of his Gesetz 
must judge for himself, and will, at any 
rate, be certain to learn much from the 
long array of examples by which each ‘law’ 
is illustrated, even if the exceptions seem 
to him too numerous to allow of such a 
term at all. 

The value of a book like this is not to be 
gauged by the amount of conviction which 
it produces. It is much to be able to show 
(for instance) that spondees at the beginning 
of pentameters are not nearly as common 
as dactyls; that an isolated spondee con- 
tained in one complete word with a pause 
in the sense after it is very rare indeed. It 
it quite another thing to be told that these 
inductions from a number of instances con- 
stitute a ‘law’: at best they can only be 
considered guiding rules. 

I will now touch on some points in which 
I feel myself to be in direct antagonism 
with the views of Hilberg. Some of these 
relate to the Jbis. There are two penta- 
meters in which MSS. agree to place 
quamuis after the verb to which it refers, 


45 Non soleant quamuis hoc pede bella 
geri. 

58 Non soleant quamuis hoe genus ipse 
sequi. 


In the second of these, G (the Galeanus) 
gives Quamuis non soleam, against metre. 
Hilberg, full of his ‘law’ C, that the 
natural order is to be looked for, seizes on 
this fact, and transfers G’s quamuis non 
soleam to 45, writing then Quamuis non 
soleant. With every wish to give G its full 
weight as a unique testimony, I confess 


this seems to me unjustified and improbable. 


239, 240 
Flebat ut est infans fumis contactus amaris 
De tribus est cum sic una locuta soror, 


Why not cum sic de tribus est ? Probably 
because de tribus counts as a single word. 
Hilberg allows this: yet, obedient to his 
law of natural order, cannot believe that 
the poet wrote anything so inverted as De 
tribus est cum sic and substitutes ¢wm. To 
me this spoils the verse, and if it were so 
written in the MSS. I should have felt its 
unusualness and un-Ovidian character, and 
should have been inclined to alter it into 
cum. And will not most of my readers 


agree with me if I find in the position of 
M 


162 


De tribus at the beginning of the first half 
of the line a designed antithesis to una at 
the beginning of the second? 


96 Qui scit se factis has meruisse preces. 


Hilberg denies that se scit, which is found 
in several of the best MSS. including the 
Turonensis (T), can be right. Why? 
Because there is no emphasis on se and 
because scit se is the natural order. I 
cannot feel this to be at all convincing : and 
I notice that Mr. Housman,! the most 
recent editor of the poem, prints se scit. 

Two cases, where I have been the first 
to make a conjecture subsequently made by 
others, must not be passed over in silence. 
One is in A. A. ii. 307, 8 :— 


Ipsos concubitus, ipsum uenerere licebit 
Quod iuuat et quaedam gaudia noctis 
habe. 


On p. 393 of the American Journal of 
Philology for 1892 I emended this: et quae 
clam gaudia noctis habet (or habes), con- 
structing Quod iwuat with ipsum. ‘ licebit 
uenerere ipsos concubitus ipsamque uolup- 
tatem coitus et gaudia ueneris quae tacet 
(taces).’ Hilberg has made this identical 
emendation, p. 653 of his Gesetze (published 
in 1894), punctuating however with a 
comma after licebit and writing the penta- 
meter 


1 In fase. 2 of Postgate’s Corpus Poetarwm Latin- 
orwin. 


THE CLASSICAL REVIEW. 


(Quod iuvat et quae clam g. n. habes. 


which I do not quite understand. 
The other is in Pont. ii. 7, 23, 4 which 
the Bavarian codex gives thus :— 


Crede mihi, si sum ueri tibi cognitus oris, 
Nec planus nostris casibus esse potes. 


In a review of Korn’s edition of the 
Pontic Epistles which I published in the 
Academy of Jan. 8, 1870 I emended this :— 


Nec planus (an impostor) e nostris casibus 
esse puter. 


This conj. I subsequently sent to Merkel, 
who admitted it into his text of 1884. In 
Giithling’s edition of the Pontic Epistles a 
nearly identical conj. Nec planus in nostris 
casibus esse putor is admitted, of which 
planus in is ascribed to A. Rothmaler, putor 
to Korn. On _ purchasing Rothmaler’s 
pamphlet I found that he read the line Nec 
planus in nostris casibus esse potest: but 
that his work was not published till after 
the battle of Sedan, Sept. 1, 1870. 

These facts are completely misstated by 
Hilberg and he even seems to believe that I 
meant puter to be an adjective. Of course 
I meant it to be the pres. subj. of putor: 
but it is perhaps improbable that Ovid 
should have combined sum with puter: 
then, with S. G. Owen in the new volume 
of Postgate’s Corpus P. L., I should read 
putor. 

Ropinson E1.is. 





EDITIONS OF CLAUDIAN BY BIRT AND KOCH. 


Monumenta Germaniae Historica. Auctorum 
antiquissimorum tomus X. Claudit Clau- 
diani carmina recensuit THEODoRUs Brrr. 
Accedit appendix uel spuria uel suspecta 
continens. Berolini apud Weidmannos. 
1892. Pp. ecxxx. 611. 30 Mik, 

Claudii Claudiani carmina recensuit JULIUS 
Kocn. Lipsiae in aedibus B, G. Teubner. 
Pp. Ixi, 346. 3 Mk. 60. 


Aw adequate critical edition of the last of 
the Latin poets has long been desired ; and 
it is a tribute which he well deserves. For 
although his graces are often of the engine- 
turned order and his lustre metallic, yet in 
power and range and deftness of poetical 
expression he recalls the best ages of Rome 
while the purity of his Latinity and his 
mastery of metre, wonderful in any case in 


a foreigner, are almost miraculous in a 
successor of Ausonius. It is however to 
his historical importance, not to his poetical 
merits, that we owe the editions now to be 
passed under review. 

The first of these editions is in scope and 
method a truly ‘monumental’ work, indis- 
pensable to every student of Claudian. Of 
its mere contents even it is not easy to give 
an adequate account within the limits of a 
review. The two hundred and thirty pages 
of prolegomena each containing forty-seven 
square inches of print as against forty-two 
in this article, to say nothing of the -differ- 
ence in type, deal with every topic in which 
any reader of the puet may be supposed to 
take an interest. Its divisions are as 
follows. I. On the life and writings of 
Claudian and contemporary history (‘tem- 


é Gow,’ 


nm 
3 
i.) 
a 





THE CLASSICAL REVIEW. 165 


porum historia’); fifteen sections and a 
chronological summary, pp. 1.—Lxix. II. 
On the Greek poems of Claudian, pp. uxx.— 
Lxxv. III. On the tradition of the Greater 
Claudian, ‘ Claudianus Maior,’ viz. all the 
Latin poems except the de raptu Proserpinae 
and the Panegyricus dictus Probino et Oly- 
brio consulibus, including the following 
sections: 1. On the early adventures of the 
Latin poems (‘de carminum Latinorum 
primis fatis’), pp. LXxXvI.—.xxx. 2. On 
the manuscripts of the Greater Claudian, pp. 
LXXXUI.—Cxxvill. 3. On the order of the 
poems in the Greater Claudian, pp. cxxv111.— 
cxtvI. IV. On the manuscripts of the Lesser 
Claudian (‘de raptu  Proserpinae’), pp. 
CXLVII.—cLy11I. V. On the tradition of 
panegyric of Probinus and Olybrius’ consul- 
ship, pp. CLIX.—cLXxII. VI. On the appendix 
of lesser poems (i.e. the twenty-two doubtful 
or spurious ones), pp. CLXII.—cLxx1r. VII. 
On the manuscripts of the excerpts (flori- 
legia) and on the scholia, pp. CLXxt1L— 
cuxxxu. VIII. On certain editions of 
Claudian, pp. CLXxxui1.—cciy. LX. Appen- 
dix in smaller type of Collectanea Gram- 
matica Metrica, in three sections: 1. On 
spelling, pp. CCV.—ccXxI. ; 2. sfetrical obser- 
vations, pp. CCXI.—CCXIXx.; 3. Grammatical 
questions, pp. CCXx.—ccxxy. Collation of 
the numbering of the poems with that of 
Gesner. Index Locorum to the pretace. 
The next 422 pages are occupied by the text 
and critical commentary. The tall pages 
rise in four tiers—the words of Claudian, 
the passages where others have imitated 
him, passages which he has imitated, 
variants of the manuscripts and conjectures 
of the learned. Then follow an index of 
proper names and an ‘index uerborum’ to 
the Latin poems, and another index to the 
Greek ones, and the book concludes on the 
611th or, including the preface, on the 41st 
page with a list of addenda and corrigenda. 

The multifarious and often intricate 
subjects of the preface the editor 
handles with full knowledge, yet not 
on the whole in unnecessary detail. The 
extent of his acquaintance with the con- 
temporaries and predecessors of Claudian is 
shown not only there but also in the abund- 
ance of parallel passages which stand above 
his critical annotations. The first question 
in dispute, the birthplace of the poet, is satis- 
factorily settled. All the evidence points 
to Egypt except the phrase of Iohannes 
Lydus, in a passage which is incidentally 
emended, de magistr. i. 47 otros 6 TlapAayov 
which is shown by parallels from Procopius 
and the Anthology to be a derisive, not a 


gentile appellation, <A interesting 
question is how and when Claudian obtained 
his proficiency over Latin 
would appear that he was from a boy 
‘sine dubio bilinguis’ ; and it is rather less 
than the truth that ‘non saepe oratio eius 
ab ingenio Latino abhorret ut male graeciss- 
antem audire uidearis.’ In fact the list of 
these infelicitous graecisms (p. 8, nm. 4) 
requires revision. There is no Greek 
attraction in Pan. 167 sg. ‘meminisse Probi 
quo uindice totam | uidimus Hesperiam 
fessasque resurgere gentes.’ We need not 
go to Musaeus and Homer for parallels to 
‘induere patrem’: Persius and Tacitus will 
serve. It is not credible that Claudian used 
‘rapere’ for sorbere (Ruf. i. 307, ii. 121) 
through thinking of the Greek fodeiv; in 
his time the words resembled each other no 
more than rap resembles ref in English. 
The approximate date of Claudian’s birth 
is deduced from the ‘deprecatio ad Hadri- 
anum,’ carm. 22, a poem full of 
obscurities: Hadrianus, himself an Egyp- 
tian, was ‘comes’ of the public exchequer 
in 395, ‘ magister officiorum’ 397—3S99 and 
praefect of the praetorium in Italy 400— 
405 (and again 413—416). By some 
juvenile indiscretion (‘me lubrica duxerit 
aetas’)—perhaps the epigram ‘ de Theodoro 
et Hadriano,’ carm. min. 21 ‘Manlius in- 
duiget somno noctesque diesque : | insomnis 
Pharius sacra profana rapit ’—the poet, who 
was now at Rome, had seriously offended 
the magnate, and his face had ceased to 
shine upon his protégé. Claudian, estab- 
lished, as in 397 he was, in the favour of 
Stilicho, could hardly have feared the resent- 
ment of Hadrianus, while there is nothing 
to prove either epigram or apology earlier 
than 396. In this case Claudian could 
hardly have been born before 379, The 
first public appearance of Claudian as a 
Roman poet is rightly said to be the Pane- 
gyric of Probinus and Olybrius. carm. man. 
41, 13. sqg. ‘Romanos bibimus primum te 
consule fontes | et Latiae accessit Graia 
Thalia togae | incipiensque tuis & fascibus 
omina cepi | fataque debebo posteriora tibi 
is adduced and thus explained: ‘Claudi- 
anus Probino consule primum fontes 
Romanos bibit uel moratus est per Italios, 
et Musa eius quae adhue Graeca fuerat ad 
togam Romanam accessit, uel Graecus poeta 
in domicilium transiit occidentale, Italicum, 
Romanum.’ I doubt both reading and 
interpretation. The metaphor of 7 
means that then the poet devotes 
himself to Roman subjects: “i Pro- 
pertius represents himself in nile 0 879- 


more 


verse, It 


min. 








164 


as about to drink at the spring from which 
Ennius drew inspiration for his Annals. 
In 14 for ‘ accessit,’ which cannot be under- 
stood in the way proposed, the oldest and 
other manuscripts read cessi¢ with the ex- 
cellent sense that the poet’s sportive Greek 
Muse made way or was abandoned for 
Roman national themes. If the Panegyric 
was the earliest, the de raptu was, it is ingeni- 
ously argued, amongst the earliest poems. 
It is dedicated to Florentinus who, as 
Prefect of the City (895—397), carried out 
the plans of Stilicho for securing Rome 
against famine, when Gildo rose in rebellion 
and cut off the African corn supplies ; and 
the theme of the unwritten portion was 
Ceres’ bounty ‘unde datae populis fruges,’ 
as the poet himself declares i. 50, where 
the goddess furnished the subject but the 
minister supplied the motive (prae/. 1. 50 
‘tu mea plectra moues’). Whether this 
conjecture be correct or not, it seems. un- 
doubted that Florentinus, who was cashiered 
at the end of 397, would not have been thus 
complimented by one who had then become 
the domestic poet of Stilicho. ‘The materials 
for a life of Claudian are scanty; nor is 
it necessary to follow the editor through 
his work of piecing and supplying. He 
leaves it at p. xxIv. to give a sketch mainly 
drawn from the paper by J. Koch, the 
Teubner editor, in the Rhein. Mus, 44, 575 
sqq. of the important events of 395—400. 
The tyranny and fall of Rufinus in the 
East, the defeat of Gildo, the war with 
Alaric and his Goths, not to speak of less 
important events like the marriage of the 
feeble emperor Honorius, and the rise and 
fall of Eutropius in the government of the 
not less feeble Arcadius, furnished abundant 
material to the laureate of the Western 
court, . On page XLII. we return to our 
subject and trace his poetical activity in 
unbroken connexion with the events of the 
time. His advancement culminated in an 
honour far more appropriate than a peerage 
as a reward of literary merit, a statue 
erected to him by Imperial order in the 
splendid forum of Trajan, with an inscription 
in which the speech of the Greek Thalia 
and of the Roman ‘toga’ are both called 
into requisition to honour the ‘ praeglorio- 
sissimus poetarum’ who combines in one 
BipytAtoto voov kat podoav ‘Ouypov. The last 
of the public poems of Claudian is on the 
sixth consulship of MHonorius, and was 
published in January 404. To the same 
year the editor assigns the Zaus Serenae 
(c. m. 30), a poem in honour of the wife 
of Stilicho. It lacks a prologue and breaks 


THE CLASSICAL 


LEVIEW, 


off abruptly at v. 236. As, like the Latin 
Gigantomachia where the break is in 
the middle of a line, it ends with a com- 
pleted sentence, he regards the alternative 
that the conclusion has perished as not so 
likely. The argument appears to be sound. 
Since it is not more probable per se that a 
page should have ended with a finished 
rather than an unfinished sentence, and as 
the lines in the rest of the Zaws which 
would, or might, give the impression of doing 
so, were the subsequent portions lost, are, 
according to my reckoning, about 52 out 
of 230, the odds in favour of the poem 
being unfinished are 3 or 4 to 1. For 
this and the sudden cessation of Claudian’s 
poetical activity, the editor offers the not 
unnatural explanation that both were inter- 
rupted by their author’s sudden death. 
Certainly one would be glad to think that 
the final task of the last of Roman poets 
was the celebration of the glorious victory 
at Pollentia, and that he did not survive to 
see the degradation and death of Stilicho, 
and Alaric master of Rome. But the 
hypothesis of sudden interruption can 
hardly be applied to all of the other un- 
finished works, the Raptus, the Bellum 
Gildonicum and the Latin Gigantomachia ; 
and thus the only argument that remains 
is the doubtful, if plausible, one of silence. 
The weighty authority of Gibbon is given to 
another hypothesis—that. Claudian was in- 
volved in the ruin of his patron Stilicho ; 
but this would not explain the silence of 
the muse of Claudian for the intervening 
years, which, as is well pointed out on 
p. LXx., offered several occasions when we 
might expect it would have been heard. 
On the question of Claudian’s religion, the 
editor writes ‘animo ille indifferens fuit 
circa religionem nisi quod, ubi coactus est, 
consuetudini temporum hoc condonabat ut 
christianum ritum obseruaret’; but it 
must be added that his sympathies were 
pagan, not Christian, as is shown by ec. min. 
50 whose significance Birt hardly appre- 
hends. The bantering tone used towards 
Peter, Paul and other Old and New Testament 
worthies shows one whose spirit is outside 
the pale. The question has been wrongly 
mixed up with that of the authorship of 
c. min. 32, de Saluatore, the genuineness of 
which Birt rightly defends, though line 7 
must then certainly be an interpolation as 
the Teubner editor has seen. This cold and 
artificial composition may well have been 
written to order, as hymns have been in 
later times. It seems to have been unknown 
to Augustine, who describes Claudian as ‘a 


THE CLASSICAL REVIEW, 


Christi nomine alienus’ (Ciu. Dei v. 26). 
What is most significant of all is that it 
would be impossible to discover from any of 
Jlaudian’s poems that they were addressed 
to the personages of a Christian court. 
The chapter on the Greek poems need not 
detain us. For the authenticity of the 
Greek Gigantomachia and the non-Christian 
epigrams a fair case, all that we can 
demand, is made out, while that of the two 
Christian fragments is most reasonably 
doubted, stress, as we might expect from 
the author of the history of the Latin 
Hexameter, being laid upon the metrical 
treatment. The next chapter brings before 
us the most important aspect of the book— 
its character as a contribution to the history 
and tradition of Claudian’s Latin works. 
One of the most remarkable features in 
this tradition is the threefold separation of 
the poems, the Panegyricus and the Raptus 
having come to us by different routes from 
the ‘Greater Claudian.’ Birt believes that 
the guiding principle of the editor of the 
last named was to do honour to Stilicho 
and that this was the reason of the exclu- 
sion of the two works which belong to the 
pre-Stilichonian period; from which it 
would follow that this posthumous publi- 
cation must be placed between 404 and 408. 
To this Stilichonian collection of the longer 
poems the collection of the minor poems 
was added in the course of the fifth or sixth 
century. Though thus separated, the poet's 
works were none of them entirely sub- 
merged, as was the fate of some other 
Roman poets ; and we still have a MS. of a 
portion of them written not more than 500 
years after his death, while from the eleventh 
century onward are abundant proofs that 
he was well-known and esteemed. 

For ‘the greater Claudian’ in this edition 
the whole or portions of twelve manuscripts 
including the two excerpta have been col- 
lated by the editor himself or by others, 
chiefly the former: besides this, reports 
and in many cases specimens are given of 
nearly a hundred others (s or ‘ deteriores oF 
For the Raptus and the Panegyricus the 
corresponding numbers are nine and seventy- 
five, nine and twenty-eight respectively. 
‘Besides these, reports are given of nine 
florilegia which contain extracts from 
Claudian and mention made of eleven more. 
The difficulty of controlling such an 
enormous mass of manuscript materials 
is much enhanced by the complexity of 
their relations. The descendants of the 
archetype have been crossed and re-crossed 
until their pedigree is almost undecipher- 


165 


able. Omitting minor variations, the longe: 
poems are arranged in no less than six 


different ways, and the shorter ones in five. 
And interpolation has gone hand in hand 
with conflation and _ re-arrangement. 
According te our editor it had begun in the 
copies of the ‘ greater Claudian,’ made not 
later than the ninth and perhaps as early 
as the sixth century. The pedigree of the 
authorities selected for the greater Claudian 
is thus described. In the fifth or sixth the 
‘codex archetypus perantiquus’ arose by 
the union of the larger and smaller poems. It 
was probably written continuously in capitals, 
Of this two copies were made, « and y; of 
y again two, z and w, of which z was the 
better; lastly from w was made a copy a. 
The representatives of # are V = Vaticanus 
2809 (the first thirty-nine leaves only), 
twelfth century, a source of great value, 
and P= Parisinus 18552, twelfth or begin- 
ning of thirteenth, which has a number of 
readings derived from y principally through 
w; other members of this family are G= 
Sangallensis S.n. 429, ninth century, the 
Gigantomachia only, and G = Reginensis 123, 
A.D. 1056, the de Nilo only. From w’s 
descendant a came B=Neapolitanus Bor- 
bonicus iv. E 47, thirteenth century, and 
A=Ambrosianus 8 66 sup. fifteenth cen- 
tury. From z the other copy of y came, 
R=Veronensis 163, ninth century, con- 
taining only the carmina minora, C = Brux- 


ellensis 5380—5384, eleventh or tenth 
century, E= the Florentine ‘excerpts’ 


entered in the editio princeps in the 
National Library at Florence, A 436, and 
« the cognate ‘excerpta Gyraldina’ in the 
Aldine edition, at Leyden 757, G 2, 
Il=Par. 8082, thirteenth century, once in 
the library of Petrarch ; all of which show 
contamination, but of different character 
and extent. For the minor poems, absent 
in II, one of our Trinity MSS. in the Gale 
collection—O 3.22, thirteenth century, col- 
lated by J. Jenkinson (the University 
Librarian is meant)—is employed, except 
for the Epithalamium Palladti and the Laus 
Serenae, where for some unexplained reason 
the collation is deficient.! Of these, V, 
the two G’s, C, E and e were used in Jeep's 
edition (1876—9); but new collations have 
been made for this edition. Time alone 
can show whether in all cases the best 
representatives of the different currents of 
tradition have been selected. With regard to 
the ‘deteriores,’ Birt candidly admits that 
‘intersunt qui codicem II laude aequent, 


1] have supplied the deficiency, Journal of 
Philology, xxiii. pp. 202 sqq. 


166 


uincant B’ ; but that of courseis no reason 
for including them in the apparatus unless 
their merits supply the defects of other 
manuscripts. The most surprising feature 
in the apparatus is the appearance of a 
late fifteenth century codex (A)—a pheno- 
menon which those who condemn the manu- 
scripts of the renaissance wholesale will 
do well to study—in company so much 
older than itself. That its excellence is 
derived from tradition and not from cor- 
rection, one instance is enough to show. 
With the help of this ‘ optimus codex’ as 
he calls it, Heinsius restored the true read- 
ing ‘Thebaeo’ in carm. min. 27, 91 where 
our other authorities give ‘Thebano.’ To 
know that Thebes in Egypt formed a 
different adjective to Thebes in Boeotia 
would have been strange learning in a 
renaissance scribe. But what are we to say 
when we observe that A does not give 
‘Thebaeo’ itself, but only a meaningless 
sequence of letters ‘Thabes,’ from which 
however the genuine reading may be 
at once restored? The editor takes a some- 
what lower view of the ‘excerpta’ than has 
hitherto prevailed. Like many other of 
the sources of the text of Claudian, they 
are a compound of gold and dross. Though 
with V they rank first in authority, they 
abound with interpolations, including even 
w Species,.the removal of gross expressions, 
the very existence of which Madvig has 
denied, Adversaria i. 11 note. 

Passing over the florilegia and the scholia, 
of which latter only a very few seem to be 
as old as the tenth century, as presenting 
nothing of value for the text, we have still 
other problems awaiting us. The critical 
student of the tradition in most Latin 
classics has done with antiquity as soon as 
he has passed the renaissance, and indeed 
very often before ; he need assign no more 
weight to the peculiarities of later manu- 
scripts than to the conjectural divinations of 
his contemporaries, But this is not the 
case with Claudian. It is not the first or 
the second of the printed editions of the 
poet, but the eighth, the famous ‘ Isengri- 
nian,’ published at Bale in 1534, that has 
put us in the possession of lines which 
appear in no manuscript source extant or 
elsewhere recorded. These lines, eight in 
number, JV. Cons. Hon. 315, 432, 509, 
636—7 ‘fluctus—consule,’ and Pan. 201— 
204, appear to have been derived from a 
lost codex, belonging to Wolfgang Fabricius 
Capito, the well-known Strasburg Reformer. 
These lines, though clearly in the poet’s 
proper style and supplying obvious defici- 





THE CLASSICAL REVIEW. 


encies, would naturally be regarded 
with suspicion but for the singularly 
fortunate circumstance that one of them 
itself reveals the reason of its own omission. 
For in both ZV Cons. Hon. 636 and 637 the 
same words te consule occur in the same 
place in the verse. And if the pertinacious 
sceptic should contend that the wily forger 
has fabricated even this appearance of 
genuineness, let him explain how it is that 
it has been allowed to appear in only one 
instance (or possibly two, cf. JV. Cons. Hon. 
509, 510) out of the five. 

How far back goes the line of tradition 
of which this particular MS. of Capito was 
the sole surviving heir, it is impossible to 
say. Birt thinks that the encomium on the 
IVth consulate of Honorius which is absent 
in V was also absent in # and so derived 
by P from the family of y. As we do not 
know what V (an imperfect MS. with its 
deficiencies supplied by a later hand) con- 
tained originally, this is obviously an arti- 
ficial hypothesis. At any rate, if it does 
not go as high as Q, it certainly mounted 
above y to y! as Birt calls it. 

The readings of this unique manuscript 
are to be ascertained by means of sub- 
traction. Michael Bentinus, who began 
this edition for Isengrin but died before it 
was completed, founded his text upon 
Camers (1510) plus the readings of the 
codex of Capito. ‘ Ergo,’ says Birt, ‘qui ab 
Isengrinio Camertem detraxerit aut egregie 
fallor aut habebit fere librum Capitonis ; nisi 


‘quod nonnumadtanr quamuis raro etiam alias 


editiones consuluisse uidetur Bentinus.’ 

Nor are we yet at the end. An edition 
of the seventeenth century (Claverius, Paris 
1602) must be ransacked for the genuine 
fragments of a MS. which now has dis- 
appeared. Its editor, ‘umbra magni 
Cuiacii, elegantiarum captator magis quam 
cognitor, ingenii festiui uentosi ac paene 
subdoli,’ has buried the witness of two 
‘ancient copies’ belonging to his patron 
Cuiacius under a mountain of rubbish and 
corruptions. Through the Augean task the 
Teubner editor has already waded. His 
conclusions in the pamphlet ‘De codicibus 
Cuiacianis quibus in edendo Claudiano 
Claverius usus est’ (Marburg 1889) are 
accepted by Birt. _He has shown that one 
portion of his collections consists of the 
variants which are entered as it would seem 
by the very hand of Cuiacius in the margin 
of the Gdttingen copy of Isengrin’s edition 
and are often called the ‘ excerpta Gudiana,’ 
and these are almost all found in Ambro- 
sianius M 9 sup. (M), so that M is one of 





THE CLASSICAL REVIEW, 


the codices of Cuiacius. The remainder 
come from a lost codex X, not dissimilar in 
character to the ‘liber Capitonis,’ often 
preserving the truth with V or E or C, and 
in two passages, as it would seem, alone, but 
at the same time deeply corrupted by an 
‘interpolator at once most learned and most 
daring.’ 

For the Panegyricus the authorities 
are partly the same as the foregoing, E, « 
and the two last named editions, but mainly 
different, viz. the codices W = Antverpi- 
ensis ill. 59 (12th cent.), R=lat. bibl. 
Arras n. 438 (12th to 13th), L=Lauren- 
tianus 33. 4 (13th), F = Florentinus bibl. nat. 
vii. 144 (13th), P=Parmensis bibl. reg. 
2504 (13th), B = Neapolitanus iv. E 47 
(13th), T'=Turicensis Carolinus 10 (end of 
13th or beginning of 13th). In the 
Raptus besides « (EK is absent from i. 25) 
and the editions, our authorities are W, F= 
Florentinus bibl. St. Crucis xxiv. sinistr. n. 
12 (12th cent.), D = Brit. Mus. 6042 (13th), 
A=Okx. Bodl. Auct. F 2, 16 (13th), C=Coll. 
Corp. Christi Cantab. 228 (13th), B=the 
third portion of the same codex cited 
occasionally, S= Paris. Lat. 15005 (14th), 
V = Antverpiensis N 71 (14th). 

Whether the editor has made the best 
selection from the enormous number of 
codices of Claudian, and again whether his 
collations faithfully represent the evidence 
of his selections, no one can judge who has 
not himself traversed the same ground. 
For my own part I should feel little hesita- 
tion in predicting that the answer to both 
these questions will be ‘ Yes’ and the future 
will discover little to the critical student of 
Claudian. For wherever I can test his 
judgments, as in the comparison of selected 
manuscripts and readings recorded, | find 
them marked by great sobriety and no little 
discrimination, and I receive the impression 
that in his company I cannot go far wrong. 

It is a matter for some regret that the 
editor has not bestowed upon the language 
of the preface the same consistent care 
which he has devoted to the matter. ‘The 
readers of so valuable and laborious an 
introduction to Claudian would have wel- 
comed in it a reflection of the elegant style 
and correct Latinity of the poet and they 
well may ask why, if the editor’s own book 
das antike Buchwesen is to be disguised out 
of all recognition in the classical attire of 
de re libraria, they should be led a dance 
through every style and period of Latin 
literature in ‘emortualis,’ ‘consectarium 


‘est,’ ‘prosiliuit,’ ‘indubium,’ ‘ingratitudo,’ 


‘compilator giganticus.’ In one or two 


167 


instances this absence of 
has obscured the meaning. 
The text which is built up out of these 
materials is not unnaturally conservative. 
Conjectural emendations are, as a rule, not 
admitted into the text ; and, in the editor's 
own modest words, we have in this edition 
‘Claudianus nondum emendatissimus.,.nec 
omni modo sibi redditus at potiorum 
emendationi apparatus.’ The * potiores,’ 
whoever they may be, will not do anything 
better than his ‘rudant’ for‘ ruant’ Pan, 
Manl. Theod, 300, ‘ille citas’ Eutrop. praef. 
ii. 9 for ‘illicitas,’ which have been rightly 
placed in the text. We owe him gratitude 
for taking this view of his task. It is 
well that there should be an edition to let 
us know the best that tradition or ancient 
correction can do for the poet's text. 
The world of scholarship will take some 
time to digest the ample spread before it ; 
and till then none but the most adroit 
conjectures are likely to be admitted. Of 
the textual problems a single example may 
be given. In the /aus Serenae 86 sgq. the 
vulgate is ‘nec tua mortalis meruit cunabula 
nutrix | ubera prima dabant gremiis redo- 
lentibus Horae | ternaque te (se should be 
read) nudis innectens Gratia membris | 
afflauit docuitque loqui’ and thus read the 
‘vetus Cuiacii.. But the other authorities 
disagree as follows: gremio P[{J|A, redolente 
A, aure Ee, aure V, awlé P! [J], napes A. 
Heinsius and Koch, building on A and 
partly on P[J], ‘gremio redolente Nupaeae,’ 
and this conjecture Birt accepts. It is 
supported by ‘nymphae’ S¢i/. ii, 345; and 
‘Napaea’ occurs in Venantius Fortun. vi, 
1, 105 ‘quod carmen scatet Claudianeis,’ 
while again Nymphs and Graces are often 
associated. On the other hand ‘aurae’ and 
the like are most naturally explained from 
‘horae,’ the two words being elsewhere 
confused in MSS,, the sense of which would 
have an excellent parallel in Pind. Pyth. 9, 
60, while for the union of the Hours and 
Graces we may compare Hymn. Apoll. 194. 
It may be that the corruptions of the last 
word point to a confusion in capitals of 
AVRAE and NAPAE. But one thing is 
clear whether we read ‘ Napaeae’ or 
‘Horae’: the reading we reject must be 
assumed to have arisen from deliberate 
unnecessary and learned interpolation or to 
have existed already in the unrevised 
autograph of Claudian. Accordingly, 
though Birt has retained a good many 
corruptions in the text which he would 
have been justified in removing, this can 
hardly be said to detract from the merits ol 


* sprachgefiihl ’ 


168 


‘his work. The punctuation has been 
thoroughly revised and in many instances 
improved. 

‘To pass to the testimonia. The reader 
should be very grateful for such an 
ample collection of parallel passages, and 
especially for those which he owes to the 
edito1’s familiarity with the verse which 
was written by our author’s contemporaries. 
But I am afraid that all the same he will want 
to know about many of Birt’s ‘testimonies.’ 
What are they evidence of? A critical 
‘testimonium’ does or should mean some- 
thing perfectly definite, a quotation from 
another writer containing a thought or 
expression the direct offspring or parent of 
the one in the text. How then, will he 
ask, are two of three quotations on p. 
186 ‘testimonia’?—Pan. Theod. 270 ‘nuntia 
uotorum celeri iam fama uolatu | mouerat 
Aonios—lucos,’ Virg. Aen. iii, 121 and 
274, Mart. viii. 65 ‘redimita comas.’ At 
this rate we might have half the Latin 
literature and dictionaries copied out as 
‘testimonia.’ If again in v. 270 Claudian 
was following Virgil and not simply using 
a poetical commonplace, the ‘testimonium ’ 
to cite is Aen. iv. 173 sqq., not Aen. ui. 121. 
In these examples the resemblance is real 
though unimportant, but what resemblance 
there is between Pan. Yheod. 332 where 
Rhaetia boasts it is the parent of the Rhine 
and the Danube, and Prop. iv. 10, 41 wherea 
Gaul boasts of his descent from the Rhine, it 
is impossible to see. The editor himself ina 
s2cond revision would, I am sure, have seen 
the inadequacy of such comparisons. 

With the view of giving a better idea of 
the text I subjoin a comparison of the chief 
differences between it and the last critical 
edition (Jeep, 1876-9) in the Raptus book 1. 
I give Birt’s reports of the MS. readings 
where, as often, they differ from Jeep’s :— 

i. 6 totum MSS. B(irt) K(och), solwm 
J(eep); Slimina BK, culmina J ; 16 leuisque 
BK, laetusque J (so AFV); 21 opibus BK, 
foribus J (ex coni.) ; 35 maritti BK, maritae 
J with the ‘ vetus Cuiacii’ ; 59—63 repune- 
tuated BK; 61 certisque BK, certis J; 
67 wia alle pepercit BK, uix lla: pepercit 
J; 71 getica BK, gelida J (so F); 98 
uanas BK, wacuas J with F; 99 grati quod 
BK with S (supert quod W), quod grati J 
(so FDACBYV) ; 103 thalamis BK, thalamos 
J, so FW(S); 113 dicto BK with ACBI, 
dictis J (so FSWDYV); 140, 141 bracketed 
BK, J adopts Baehrens’ emendations ; 143 
magna B (text) with W, cuncta B (conj.) K, 
una J with the others (ABC omit 139— 
213); 159 Vellitur B (comparing ii. 143) 


THE CLASSICAL REVIEW. 


K, Vertitur J (Cuiacii mg.); 164 motibus 
BK, molibus J; 165 nutrit BK, mittt J ; 
171-8 bracketed by Jeep; 172 amnis BK 
with D, cgnis J (as the rest); 174 Offenso 
rimosa BK, Offensus rimata J with F; 195 
cui B with 8, tibi JK (as the rest); 213 
intendit BK, ext. J with F; 244 westit B 
(comparing Lucan x. 119) with 8, cingzt 
JK ; 288 spectantes BK, sperantes J. 

In the majority of these passages the 
reading of Birt’s text is better or at any 
rate not worse than its rivals ; and it may be 
defended either as extrinsically or else as in- 
trinsically probable ; but it is clearly wrong 
in two cases. in which the tradition is re- 
tained : 67 where Jeep well compares Val. FI. 
v. 453, and 164 where motibus is intolerable ; 
and in others it must be felt to be doubtful. 
Thus in 99 it rests on a theory of the 
limitations to the use of spondaic words in 
the fourth foot of the hexameter, prolegomena 
p. 214, which has not been properly worked 
out. 

The index of proper names is not a mere 
register, but comprises a good deal of 
valuable historical and geographical inform- 
ation. The index of words is fuller than 
Gesner’s upon which it is based, but not 
however complete. The usefulness of both 
would have been much increased if the 
editor had abandoned the numeration of 
Gesner, numbered the poems in the order 
of his own edition and made the references 
to correspond. There are of course a certain 
number of misprints scattered throughout 
the book, but those I have noticed are too 
unimportant to record. 

The Teubner Editor has been careful to 
avoid trenching upon the ground of his 
predecessor. The consequence is that we 
have lost one of those very convenient texts 
with short critical apparatus which we now 
expect from Messrs. Teubner. This retro- 
gression must have been regretted even if 
it had been necessary. But where was the ne- 
cessity? The appearance of Schenkl’s Auso- 
nius in the series of the ‘ monumenta’ did not 
banish the critical footnote from the pocket. 
edition of Peiper; and if Koch felt, as was . 
very natural, scruples about making full 
use of the materials collected by his friend, 
there was no reason why the Teubner 
Claudian should not have waited a little. 
Critical ‘annotationes’ there are, but 
except a few signposts they are placed 
after the preface ; and consist of comments 
of various length upon debated passages, the 
most elaborate being the discussion of the 
confusion of ‘ Hennaeus’ and ‘ Aetnaeus’ 
Rapt. i. 122. The editor has already done 


THE CLASSICAL REVIEW, 


good service to Claudian, as indeed we have 
seen in the course of this article. For the 
present edition he has ‘employed’ (adhibui) 
two manuscripts ‘ad deteriorem gregem 
plane adgregandos,’ cod. Rodomensis 1040 
saec. xiii. and Cod. Mus. Brit. Egerton n, 
2627 saec. xii. (?), neither of which adds 
anything to our knowledge, of whose 
readings he gives specimens in the preface. 
A real contribution is the conspectus of al/ 
‘Isengrinian’ readings not derived from 
Camers eacept ‘orthographica’ (praef. pp. viii. 
sqq.); but why were these omitted! Koch’s 
text, as will have been noticed above, agrees 
in the main with that of Birt from which it 
differs chiefly in the higher value set upon 
the Vaticanus. When he leaves his prede- 
cessor, he is generally in the right. Thus 
in Bell. Gild. 54 alerent (alerem B), 69 praedaC 
(praedo B), 130 Cybele sicco (sicco C'. B),[ 347 
Seruati | Dewicti coni. B), 402 Hic (inc B), 
414 dum (cum B), 441 arma oneri (arma 
ouium B). But not (to omit doubtful cases) 
in 247 petisset ( petissent B) and 395 occidit 
(obtruncat B), where the reasons alleged for 
the changes reveal a somewhat superficial 
acquaintance with grammatical and lexical 


niceties, apparently the editor’s weak 
point. Compare uf. i. 230 ‘non coniunx, 


non ipse simul, non pignora caesa | sufficiunt 
odiis, where Birt has caesi most rightly 
though Koch thinks it against ‘grammatices 
regulas.’ In B.G. 414 (above) he writes ‘dum 
contra VP qui praebent cum recepi; illud 
enim semper, hocrarius indicatiuum postulat.’ 
dum is right, but this is not the reason. 


. 
‘ 
169 


The following certain or attractive emenda 
tions of the editor’s own are among those 
which now appear in the text; Pan, 22 ‘ per 
aethram’ (for Arcton). Rufin. ii. 270 Clam 
(lam). Bell. Gild. 299 patre remoto (carcere 
moto). c. min, 32.5 numen (mundum). Rapt. 
li, 23 sq. ‘summa peremptus | ima uiget parte 
emoriens et parte superstes’ (“Ima parte uiget 
moriens et p.s.’ vulg.) In the index of 
proper names the reader is inconvenienced 
by a repetition of the error of Birt; but 
taking the edition as a whole it may be 
recommended as a serviceable and judicious 
one, 

I conclude with a few words on the 
carmina Graeca. Birt has made excellent 
use of the valuable work of Koechly and 
others on the corrupt Gigantomachia nor has 
he failed to do something himself. 1.7 ds 
kat vov 67 PoiBe seems better than anything 
else proposed ; 22 dxpw for dxriva is a fine 
correction ; 64 y6éviov for dotvwv may be 
right, but there are other corruptions as 
dvacxonevn, Which should perhaps be dvacye- 
pevat, The Teubner editor only contributes 
the remarks that in 6U xarévavre is defensible 
because it is used in the LXX., and that civ 
Tos Te Kepavvois may be tolerated in a poem 
‘serae graecitatis.’ Though he accepts 
Birt’s rervdds in 68 for te zupds, it is 
probably wrong, as is his érervevoeas (for 
-av) in 17 where we shonld take €rawot 
from Koechly and Schenkl; something 
appears to have been lost before v. 16 in 
which 颒 should be a¢’. 

J. P. PostTGate. 


HAUVETTE ON HERODOTUS. 


Heérodote, historien des guerres Médiques, par 
_ Amétp&e Havvette. Paris: Hachette et 
Cie. 1894. 10 fr. 


Ir is not easy within the compass of a 
column or two to write an adequate notice 
of a large volume, which to a considerable 
extent is occupied with the discussion of 
uncertain and disputed points. M. Hauvette 
has made a real contribution to the study 
of Herodotus; he has brought together all 
that is known about his life and travels ; 
he deals with the nature of his book, and 
the question whether or not it was finished ; 
_he passes in review the criticisms of writers 
‘ancient: and modern on Herodotus, and 
after thus dealing with the book as a 


whole, enters on the criticism of the account 
given of the Persian Wars. 

A few points may be noticed. In refusing 
to accept (as Stein does accept) the year 
468 (Euseb.) as the /loruit of Herodotus in 
the sense that Herodotus was forty years 
old at that time, Hauvette ‘is probably 
right, For if Herodotus were forty years 
old in 468, his birth falls in 08, and he 
would be a contemporary of the Persian 
Wars, eighteen years of age at the battle of 
Marathon, and twenty-eight at the battle of 
Salamis. This is highly improbable ; had 
he taken a personal part in those great 
events, or witnessed them with his own 
eyes, he would have let us know, and he 
would have been able to ascertain with 


170 


certainty facts which he leaves uncertain, 
as for instance the precise facts of the 
return of Xerxes to Asia, about which he 
ean only argue on grounds of probability. 
The difficulty of explaining the relations in 
which Herodotus stood to the Persian 
authorities—how could he who had assisted 
in expelling a vassal king of Persia from 
Halicarnassus travel in Egypt and Baby- 
lonia !—is treated with much ingenuity by 
Hauvette, who by supposing that Lygdamis 
was expelled before 454, when Halicarnassus 
was a member of the Delian League, is 
compelled to put the travels after the 
rebellion. The truth is that Herodotus in 
his history, which is our best source of 
knowledge, shows an _ admiration for 
Artemisia which is a little inconsistent 
with his sympathy with icnyopia. Perhaps 
this divided feeling may explain both the 
favour which he received from the Persians, 
and the dislike with which he was regarded 
at Halicarnassus, 

It was the opinion of Niebuhr that the 
current story of the Persian War, except in 
the broadest outlines, was quite untrust- 
worthy. He supported his view on evidence 
which combines the tradition of Plutarch with 
the statements of Herodotus. M. Hauvette 
carefully distinguishes the narrative in 
Herodotus from later exaggerations and 
thus answers Niebuhr at least in part. 
Again Niebuhr thought that the account of 
the Persian army in the Seventh Book was 
taken from a poetical source (when shall we 
hear the last of these poetical sources ‘), 
and went so far as to indicate the precise 
source. It was Choerilus of Naxos from 
whom Herodotus copied. But M. Hauvette 


THE CLASSICAL REVIEW, 


shows that Choerilus was later than Hero- 
dotus, and if there was any copying, the 
poet copied from the historian. In fact 
Niebuhr made the mistake of bringing 
Herodotus too low by twenty years or more. 

The difficulties attending Nitzsch’s view of 
the Adyor on which the history of Herodotus 
is based are well stated. . Nitzsch regards 
these Adyo. as oral, and preserved by 
memory, but reduced to a definite form, and 
in a sense literary. These Herodotus 
merely embodied in his work, as he found 
convenient. M. Hauvette shows that what- 
ever the nature of the Adyo. may have been, 
Herodotus did not embody them wholesale 
in his work. For instance, Nitzsch would 
divide the account which Herodotus gives 
of Miltiades into a Philaid source, favour- 
able to Miltiades (Bk. vi. down to e. 115) 
and an Alemaeonid source, hostile to Milti- 
ades (c. 123 f,). But in the earlier chapters 
of Book vi.—the supposed Philaid source— 
nothing is said of the conduct of Miltiades 
at the Danube, or of his conquest of 
Lemnos, facts highly creditable to him, 
which we find elsewhere in the history— 
the second of them in vi. c. 136 ff., z.e. in 
the supposed Alemaeonid account. 

In the same careful manner the theories 
of Wecklein, Sayce, Panofsky and Traut- 
wein are discussed, no less than the special 
history of the Ionian revolt and Persian 
War. ‘The book is in fact a complete intro- 
duction to the study of Herodotus so far as 
he is the historian of the Persian Wars. 
On the oriental history M. Hauvette does 
not enter. 

EVELYN ABBOTT. 





HADLEY’S EDITION OF THE HECUBA. 


The Hecuba of Euripides, with Introduction 
and Notes, by W. 8. Hanpuey, M.A. 
Cambridge, University Press. 2s. 6d, 


In the preface to this edition, Mr. Hadley 
says: ‘I have endeavoured to form an 
independent judgment on each question 
before referring to the notes of others, in 
the belief that a fresh point of view is more 
likely to be attained by an editor, who does 
not at once fly to the assistance of his pre- 
decessors, when wishing to explain a diffi- 
culty or illustrate a view.’ Freshness and 
originality are to be expected in an edition 


prepared on this plan by an editor of Mr. 
Hadley’s insight; but there is a danger 
that the commentator who makes up his 
mind before consulting his predecessors, 
may, in cases where decision is difficult, not 
only be unduly biased in favour of his 
own views and those of his predecessors 
who have arrived at those particular views 
before him, but that, from a lack of interest 
in what others have thought, he may some- 
times remain in ignorance of a view which 
after all is the right one. Both the advan- 
tages and the disadvantages of this method 
of proceeding may be seen, I think, in the 


THE CLASSICAL REVIEW, 


book before us. It contains fresh sugges- 
tions of value both as to reading and as to 
interpretation, but I have found myself in 
reading it not only often at variance with 
Mr. Hadley in his decisions—that is very 
likely due to bias on my own part—but in 
doubt as to whether he has fully and fairly 
surveyed and weighed the evidence that has 
been brought in support of rival views. 

Under the head of valuable suggestions 
here made, I may mention the reading 
mAovoiowst for the MSS. zAovcios év at v. 
624, éreoraicwpey for ereorécwope (not 
adopted in the text) at. v. 1042, AWBas Avpas 
t for AdBav Avpas (also not adopted) at v. 
1074, and the adoption of iwu7rerys, the 
reading of L, at v. 1101. 

At v. 481 the punctuation here suggested 
has been already suggested and printed in 
text by Weil. Atv. 847 Mr. Hadley reads 
Ts, avaykns for tas dvayxas. This is neat, 
and is certainly a great improvement on the 
scholiast’s suggestion that kar’ dvtimtwow 
we are to understand that when Eur. said 
Tas dvayKas ol vopor dudpicav he meant at 
dyaykat TVs vopous Owwpicav. I prefer though 
here to put a colon after cupzitve, to take 
dvaykas as ‘blood-relationships,’ and to 
interpret dipicav in the light of the fol- 
lowing words as ‘ force asunder’ (scholiast 
ducryoav). At v. 1185 f. Mr. Hadley 
writes zoAXai yap éouev? ai pev cio’ éripOovor 
(‘exposed to bad feelings’) ai & cis dpibpov 
ov kaxov Tehixapev. There is a good deal of 
mending here. I think ‘ ending’ is better. 
On v. 588 Herwerden’s dvayxatow for dvay- 
katov should at least have been mentioned. 
In the account of the MSS. Or. has by an 
oversight been omitted from the list of 


plays contained in @. 


In the explanatory commentary there are 
many good notes: e.g. at v. 309 on the dat. 


with dévos (‘akin to the dat. with déxouau’), 
at v. 353 (‘ éyovo’ =‘ causing’), v. 504 a 
suspected line is well defended by comparison 
of Soph. Phil. 343, at 614 (ri yap rab; 


‘what else can I do?’), at 774 (on the 
Thracians, one of several good historical 
notes), at 793—797 (a good defence of an 
almost desperate passage): ‘ Often have we 


‘sat at the same table; our hospitality he 


has shared more frequently than any other 
of our friends: yet, though he has experi- 
enced such kindness at our hands, he has 


slain and robbed of burial our son,’ though 


the ei xraveiy éBovAero cannot be said to be 
satisfactorily explained (‘assuming that 
there might be some reason for the desire 
to kill’), at 854 (subj. to aveiy is dixy’), at 


901 (jovyxov predicatively with pevew [agree - 


171 


ing with pe understood]), at 1058 (‘ war’ 
ixvos following on their track, éx} xeipa with 
motav ‘in which direction?’), at 1177 J 
heartily agree with the admiration expressed 
for Eur.’s ‘gallery of female characters, 
unsurpassed till Shakspere came.’ 

In the following cases I beg leave to 
suggest a view more or less divergent from 
that expressed by Mr. Hadley, or else to sug- 
gest the need of a supplementary note. At 
vv. 209 f. I do not think é6a VeKpov pera 
Tadawa Ketoopat is properly explained by the 
note ‘ her marriage will be with the shades’ : 
at v. 221 zpos épGov yay’ should, I think, be 
explained—by a comparison of zpds tiBov 
v. 261, and zpos torov apByoe v. 1263 (‘ up 
the mast, not, as Mr. Hadley, ‘by the 
mast > and Hom. € 1 zpoceBy TpHX€lav ATapTov 
climbed the path,’ cf. ‘to breast the hill’— 
as ‘up (on) the high mound’: at vv. 234 ff. 
Prinz’s little note ‘«i=nuwm’ has always 
seemed to me to let ina flood of light on 
the passage, and yet Mr. Hadley only men- 
tions the interpretation as a_ possibility, 
after a lame translation which takes «i 
as = ‘if’: at v. 398 dws is surely best 
taken, as at Z’ro. 147, as = ic@’ drws, 441— 
443 seems to me rightly rejected by most 
editors, and might well be followed by 
435—437. I much doubt (v. 441) the 
admissibility of és: = ovrws in tragedy: at v. 
461 Mr. Hadley translates aédivos dyaApa 
Awds ‘the pride of her Zeus-born son,’ Weil 
‘le monument de Venfantement du fils de 
Jupiter’: surely the latter is right: v. 490 
Mr. Hadley makes a desperate attempt to 
defend against Nauck’s suspicions. He 
translates doxotvras Sdauovwv elvar yevos 
‘ seeming to be a race of gods indeed.’ Even 
if the plur. dox. after oe can be defended, the 
sense arrived at thus is not satisfactory: 
the question is not as to the existence of 
gods but of the attention they pay to man: 
on v. 599 the note is inadequate. No other 
sense seems to me possible (in their con- 
nexion with what precedes and follows) for 
the words dp’ of rexdvres duahépovow 7) tpopat ; 
except ‘are one’s parents then of more 
importance than one’s nurture Atv. 620 
Mr. Hadley rejects Porson’s coupling of 
kddduora (adv.) with evrexvorare and all 
he can suggest in its place is to supply réxva 
with both zAcora and xdéAAurra (adj.): at v. 
828 nothing is made out of digas, though 
two alternatives are suggested. Is it possi- 
ble that we ought to read refces here? 
(‘how will you repay ¢’): at v. 867 cipyover 
xpnobat py Kara yopny Tporous the note is 
‘prevent him from using. The p is out 
of place: the order should be eipy. wy xp. 


172 THE CLASSICAL REVIEW. 


tather ‘constrain him to adopt ways he 
does not like :’ at v. 888 as with the present 
reading must be taken as = ovrws (see above 
on v. 441): rather than adopt this I should 
prefer (as an irresponsible reviewer) to read 
(here and at 1.7. 603, and Tro. 726) 
either dAAws or to as yeveobw for dAX ds 
yeveofw : at 1025 ff. Mr. Hadley on dpépoas 
Prov practically says ‘ du¢€pdw never means to 
lose but always to take away, therefore we 
must here translate it to take away from him- 
self’ This reminds one of the car’ dvrimtwow 
of the scholiasts. If Euripides had meant 
to say ‘having lost your life’ (which in the 
sequel we find Pol. did not) he could have 
sald dpepOeis Biov: at v. 1032 Mr. Hadley 
says ‘ Pol.’s attack and Hec.’s defence each 
take up fifty lines.’ So they would if with 
Prinz the editor rejected vy. 1174 and with 


Nauck v. 1191, but as Mr. Hadley prints 
them they each take up fifty-one lines: at v. 
1155 dur. croduparos, i.e. ‘both of spear 
and cloak’; but even supposing the cloak 
could be called a oréAtopa why should the 
women want to take it away, and so make 
Pol. better able to fight ?: at v. 1270 on the 
supposition that some meaning must be 
found for everything in a Greek text Mr. 
Hadley translates the line, forcing év@dde to 
mean ‘in my present position’ de. a slave. 
Weil’s comment on the line is refreshing 
after Mr. Hadley’s translation, it is ‘ces 
mots n'ont pas de sens.’ 

There is a sympathetic introduction to 
the play, and an analysis, but the edition 
contains no scheme of choric metres. 


E. B. ENauanp. 


KOCK’S EDITION OF THE CLOUDS OF ARISTOPHANES. 


Ausgewahlte Komidien des Aristophanes, er- 
klirt von THEopor Kock. Erstes Bind- 


chen: Die Wolken. Vierte Auflage. 
Berlin, Weidmann: 1894. Pp. 226. 
2 Mk. 40 Pf. 


THis admirable edition of the Clouds now 
passes, after an interval of eighteen years, 
from its third into its fourth edition with 
an increase in bulk of eleven pages. It has 
been thoroughly revised from beginning to 
end, though the alterations are less remark- 
able for their importance and quantity than 
for the conscientious carefulness to which 
they bear witness—carefulness so  con- 
spicuous by its absence from the ordinary 
English school editions of the classics with 
which the minds of our youth are misguided. 
No point is too small to escape Kock’s 
vigilance, as the insertion of the v./. yAov 
in the passage quoted from Strabo on v. 597, 
or the substitution of ‘wohl’ for ¢ sicherlich ’ 
in the note on 666 show. 

Since the appearance of the third edition 
the most important contributions to the 
literature of the Clouds have been Blaydes’ 
edition and Piccolomini’s ‘Studi critici ed 
‘esegetici’ in Annali delle Universita Toscana 
vol. xvi. (Pisa 1879 fol.), and Zielinski’s 
general work Die Gliederung der altat- 
tischen Komédie. The latter is discussed 
by Kock in a preface of four pages added to 
this edition. Though he has not perhaps 
put the case against Zielinski’s view of the 


epirrhematic nature of comedy in its 
strongest form, yet the objections which he 
enumerates on pp. 6-7 are quite sufficient 
to justify him in refusing to bind down 
Aristophanes with these new-fangled fetters 
of form or to bombard his readers with the 
contents of ‘a perfect arsenal of names 
(borrowed partly from the form, partly from 
the subject matter) for things which have 
long been quite well known without the 
want of a special terminology for them ever 
being felt.’ For a system like Zielinski’s 
must be one of two things, Hither it re- 
presents the actual mould, so to speak, in 
which comedy is cast—which is incredible: 
for, to use Kock’s metaphor, ‘ das dichter- 
ische Schaffen sprengt eben alle Fesseln, 
welche die Theorie ihm anlegen mdochte.’ 
Or it is merely a method of dividing a play 
into parts: in which case it must establish 
claims to a greater convenience than any 
existing systems, if it is to drive them out 
of the field: and this Z.’s system most 
emphatically cannot do. 

Piccolomini’s Studies are frequently 
quoted, and contain many valuable sug- 
gestions. Thus v. 144 Pice. would read 
Xawpepov tov Swxparyny, on the gronnd that 
Sokrates is the subject of éuérpyce (v. 148), 
and must therefore be the person to whom 
the question was addressed. vv. 218-9 his 
re-arrangement of the lines: STP. dépe, tis 
yap otros; MA®, otmi ris kpeudOpas avip ; | 
aitds. STP. tis adtos; is certainly an im- 


THE CLASSICAL REVIEW, 


provement. On the other hand in v. 334 
it hardly seems possible to agree with him 
in assigning the words Bécxkovc’ apyovs as an 
interruption to Strepsiades: it seems far 
better to regard dvdpas...dpyovs as an inter- 
polation and read airds for ravras with 
Kock (who does not quote Pice.’s view). 
In v. 439 Kock follows Piece. (whom he does 
not mention) in regarding dreyvas 6 tT 
BovAovrat as an interpolation from v, 453, 
a special form of corruption which Pice. 
also finds in v. 266 where peréwpo. has got 
in from peréwpov in v. 264, and in v. 1010 
where zpos tovro.ow is due to zpos Tovrous of 
v. 1022. 

Blaydes is less often referred to-—less 
often perhaps than his work deserves. His 
suggestion tyAepavots oxomias v. 281 is 
quoted, and in v. 284 Kock inclines to his 
movrov otevaxovra for keAddovra. A note on 
v. 582 remarks on his statement that jvica 
is rare in comedy, that it occurs fifty-two 
times (forty-one times in Aristoph. only) 
as against eighty times in Aesch. Soph. 
and Eur. 

The introduction is reprinted almost 
without alteration, though there are a few 
additional references to recent literature. 
Thus a note on $ 12 refers to Diels’ view 
(to which attention is also called in the 
commentary, ¢.g. on v. 627) that Ar. fre- 
quently parodies the scientific theories of 
Diogenes of Apollonia. A Jong note on 
§$ 27 quotes Naber’s theory of the diacKevy 
of the Clouds, and dismisses it somewhat 
more summarily than its merits deserve: 
for example, it is hardly a disparagement of 
the relative value of the statements of the 
fourth Hypothesis to say that the bulk of 
it is ‘nichts als Vermutung’; nor is ita 
weakness in Naber’s theory that it is not 
identical with the statements of the Hypo- 
thesis by which it was suggested. Kock’s 
own theory, which in the main agrees with 
Teuffel’s, remains unaltered. 

A few new readings are printed in the 
text. v. 47 we now have dorews, attested 
by inscriptions, for the MSS. doreos. 230 
the better form xarapecéas should be read : 
vide errata. 489 zpoBdédw with Hirschig. 
530 pv for #7, which must surely. be a mis- 
print: the metre does not require the later 
form, and Kock himself in a new note calls 
attention to the ‘ remarkable’ fact that the 
only certain instances of jv Ist pers. sing. 
in Ar. occur in the Plutus: it must, 
however, be remembered that the Plutus 
is Ar.’s latest play and that Attic had by 
that time lost much of its purity. 814 
elsewhere he now prefers the form évtavdoi 


173 
which is attested alike by inscriptions and 
by the best MSS. 985 Kndecdov, for K nKetdov, 
apparently because Kydetdns edidacxe occurs 
In an inscription of the end of the fifth 
cent., of whom the poet referred to in the 
text may have been son or grandson (Kohler). 
‘ ~ . 1G Oe | , 
Fn Tuxey dy with the MSS. for Suidas’ dv. 
: igws (Reisig) is omitted and a lacuna 
marked. 1355 Kpuov for Kptov on the 
authority of Aristarchos (Lehrs), 
these readings few 
new emendations. v. 62 he would read by 
vOev0 eAovdopovpebu to avoid the break after 
the first syllable of an anapaest in the 
fourth foot. 337 dar aiOpias dvepas (or 
iepas) yapwWovs ke. to escape the difficulty 
of eira and the want of connexion between 
deptas duepds of the MSS. 376 xdvapracbact 
héperGar for kdvayxacbaocr to avoid the tauto- 
logy with kar avayknv. 1130 rudiqvac for 
Tuxelv wv, Which is certainly very at- 
tractive ; and 1418 rév yepovra rod véov ort 
kAaew for TOUS y€povras o) veous Tt kAaeuv. 
The critical appendix still appears to be too 
brief even for a Weidmann school edition. 
The commentary has been thoroughly 
revised, and much new and _ interesting 
matter has been added, of which space 


, | 
Besides 


Kock suggests a 


only allows me to refer to a little. The 
first new note (on v. 1) is pessimistic : ‘die 


Frage, wie sich die Handlung bald in bald 
vor den Hiusern des Streps. und des 
Sokrates im einzelnen abgespielt hat, scheint 
unlésbar’: but one is quite willing to 
believe that Kock has something much 
better to do than attempt to solve it. v. 332 
odpayidovexapyoxouytas is now rightly ex- 
plained as being rather = ‘wearing onyx 
rings’ &c. 377 there is an interesting note 
on verbs in -vavac in comedy. 508 on 
kataBaivwv he observes that there must 
have been steps down into the dpovrtiw- 
typvov—a fact which does not simplify the 
problem of the stage arrangements. 923 
he emphasizes his view of Tpwrn (Welcker’s 
reading for MSS. zpdrovs) ‘ zuerst, d.h. vor 
allen andern Komiidien,’ and quotes a number 
of other compounds with dva- denoting 
‘again’ to support dvayetoa: = ‘ wieder kosten 
lassen.’ 541 6 A€ywv ray is now explained 
as meaning the protagonist. On vv. 653-4 
he adds a remark which, if borne in mind 
as a principle of general application, would 
save much perversion of ingenuity in the 
explanation and emendation of Ar. : ‘die 
Worte sind nicht sehr verstiindlich, erhielten 
aber wohl thre Erklirung durch die Gebér- 
densprache’ (the italics are mine). In 870 
he now takes zp(Bwv as referring to the 
‘ yell-known article of clothing of poor men 


174 


and philosophers.’ 880 Naber’s ovxivas for 
oxutivas is quoted and rejected on the 
ground that Herod. i. 194 mentions real 
rAota oxvtwa and therefore no change seems 
necessary—a quite insufficient ground for 
rejecting what appears to me to bea certain 
correction. 1036 he notes that the “Ad:cos 
uses a pseudo-Sokratic dialectic, while the 
Aixatos explains his principles in a connected 
speech—which, if it is anything more than 
a coincidence, would tend to show that in 
this part of the play at all events Ar. 
distinguishes between Sokrates and the 
Sophists. On 1096 there is an interesting 
note calling attention to the climax in 
1089-1100 : cvrijyopo (the lowest grade of 
official), tpaywdot (who rank next in public 
estimation), dynydpor (the representatives 
of the sovereign djpos), and finally even 
the sovereign ojos itself (=Geara/)—all 
alike are etpizpwxto.. 1194 an unnecessary 
attempt is made to defend drahAdrrowro in 
the sense of Sua\AdtrowrTo, which, if the mean- 
ing required is ‘reconcile,’ must certainly 
be restored with Hirschig : but there seems 
to be no reason for abandoning the old and 
natural explanation of Ernesti: per compo- 
sitionem aut alio modo se expedire e contro- 
versia. 1236 his suggestion that ér is used 
in its threatening sense ‘a time will come 
when’ &c. seems undoubtedly right. - 1504 
Reisig’s é£apOets = érapbets (for MSS. épac Geis) 
is defended by Zhesm. 981 éEaipe durAjv xapw 
xopeias. 1315 évyyevntat is taken = ‘ feindlich 
zusammentreffen, for which, as he admits, 
there is no parallel : but surely it need mean 
no more than ‘beat every one be has any- 
thing to do with.’ The commentary is 
enriched throughout by many new parallels 
from the New Comedy, and by frequent 
references to Guhl and Koner’s sixth ed. 
There are one or two things that one 
would like to see altered in this new 


THE CLASSICAL REVIEW. 


edition. v. 824 Dawes’ édcées is still 
read : but as the canon Davesianus (to which 
Kock adheres strictly in the constr. of od 
py also) is not borne out by inscriptions, 
and as Goodwin has given a very good ex- 
planation of the origin of the aor. subj. 
after drws pi, the change, slight as it is, 
seems unnecessary. 1149, it is surely impos- 
sible that év refers to viov and not Adyov. 1271 
Kakas ap dvtTws ecxes the imperf. appears to 
be of the kind nicknamed ‘ panoramic’ so 
common in Plato (cf. quanta laborabas 
Charybdi), and to have no reference to the 
time when Amynias. borrowed the money. 
1277 there does not seem sufficient reason 
for abandoning the MSS. zpooxexdAjoPar : 
the use of the perf. was probably suggested 
by ceccioGar in the line before. 1342 a 
note is wanted on the rare use of wore 
c. indice. in oratio obliqua, 1384 ¢pacas, 
the reading of EFG, which is mentioned 
with approval in the note, should certainly 
be printed in the text. 1431 Kock still 
approves of G. Hermann’s éz’ ixp/ov, but in 
the face of Pollux’s remark (x. 156) érevpov 
dé, ov Tas evorktdlas OpviOas eykabevdoe ocrpPeE- 
Byxev, ’Apirtodavys Neyer Oo7ep Kal Kpepabpay, 
év tais NedéAats, which can scarcely refer to 
any passage but this, and of Rav.’s «ami 
m\eiov, it is more probable that the word 
zérevpov in some form or another is concealed 
in the line: at all events the facts are worth 
mentioning in a note. 

It is impossible to read such a model 
edition as this without wishing very strongly 
that Kock would not confine his editorial 
labours to the four plays which have already 
been before the public so long in the Weid- 
mann series, but would go on and produce 
a complete edition of Aristophanes, which 
no one is so well qualified as he to under- 
take. 

F. Artaur H1rrzet. 





SIHLER’S EDITION OF THE PROTAGORAS. 


The Protagoras of Plato, with an Introduction 
and Notes, by E. G. Srauer, Pa.D. 
New and revised edition. Harper & 
Brothers. 1892. 


Tuis new edition offers a clear attractive 
page, and has the excellences of the former 
one; among others, a body of notes con- 
taining a large amount of valuable illustra- 
tive material. The changes are inconsider- 


able. Some errors in accent and breathing, 
however, have been brought over from the 
first edition. Also in the Greek index, p. 
135, both editions derive ovveoréov from 
cvvicracbar, where the notes now give the 
matter correctly. The English words 
‘would’ and ‘should’ are interchanged too 
often perhaps even for a Greek text-book, 
eg. p. 95, ll. 25 and 37; p. 115,175 
Lob sieges. Cee o ie 2. SLVEES 








THE CLASSICAL REVIEW. 


The book in the introduction and notes 
sets forth with some fulness the line of the 
argument, yet fails to make sufficiently 
clear to the student its real character and 
aim—a matter of much importance, where 
the dialogue has so much keenness and 
subtlety. Several points may be mentioned 
which the editor has failed to make clear. 
(1) The contrast between Socrates’ and Pro- 
tagoras’ ideas of dperyj, the one being 
simple — although definition is disclaimed— 
the other elastic and indefinite, aiming to 
be practical, whatever that may mean (cf. 
318E). (2) The fallacies in the arguments 
of Protagoras. For instance, as his defini- 
tion of dpery, he reaches, instead of a 
definition, that which ordinary people think 
to be dpery. (3) The unity, from Socrates’ 
point of view of the whole dialogue, in 
spite of the repeated dramatic breaks. 
Socrates does not expect an immediate 
definition of dperj, and at the close dis- 


175 


avows having attained to a definition, Yet 
again and again, by different lines, he 
directs the thought unceasingly towards 
this goal, every time forcing the conclusion 
that dpery is knowledge. (4) Socrates uses 
a twofold method of argument; serious, 
and ironical or playful,—we may call it by 
different names. In this latter, Socrates 
does not hesitate—and this with entire 
propriety—to imitate Protagoras, and to 
foil him with his own weapons. His use of 
Simonides’ poem illustrates this, and to the 
reader of to-day is delightfully sophistical, 
If the weapon is a good one, Protagoras is 
fairly beaten ; if the weapon is poor, Pro- 
tagoras’ weapon is a failure. At all 
events, when Socrates has worsted his 
opponent, he himself says the weapon is 
worthless, throws it away (cf. 347), and 
takes one more to his mind, 


J. A. Towte. 


DE MIRMONT’S APOLLONIOS ET VIRGILE. 


Apollonios de Lhodes et Virgile, by H. DE LA 
VitteE DE Mirmonr. (Annales de la 
Faculté des Lettres de Bordeaux, 1894, 1. 


pp. 1-83). 


Axout two years agoa translation of Apoll- 
onius Rhodius with commentary by M. de 
Mirmont was noticed in this review, and, 
pursuing the same subject, this dissertation 
is intended as an introduction to various 
studies on Apollonius and Vergil. The first 
of these has lately appeared, and is a for- 
midable thesis of 778 pages, presented to 
the Faculté des Lettres de Paris, upon the 
Mythology and the Gods in the Argonautics 
and the Aeneid. 

The object of the present little work is to 
define with more precision than has been done 
hitherto the influence of the Greek poem 
upon the Aeneid. The parallel passages 
in the two poems have often been collected 
both in ancient and modern times. M. de 
Mirmont gives some account of these, 
beginning with Servius and Macrobius. In 
modern times Orsini (Fulvius Ursinus) in 
1568 collected all the passages of Vergil 
which he thought imitated from Greek poets 
and some prose writers, and judiciously did 
this without making any criticisms. He has 
been followed by Tissot and Eichhoff in 
France, whose works both appeared in 1825. 


Before this however the elder Scaliger in 
the fifth book of his Poetice, published post- 
humously in 1561, had compared the best- 
known passages with the object of exalting 
Vergil and depreciating Apollonius. This 
was answered in 1641 by Hoelzlinus, who 
edited Apollonius and undertook his de- 
fence against Scaliger. As neither of them 
thought it necessary to give reasons fo1 
their preference of either poet, their differ- 
ences need not detain us, but upon the 
whole Apollonius suffers less from the in- 
discriminate abuse of Scaliger than from 
the clumsy eulogy of the Dutchman. No 
collection of parallel passages however is of 
any critical value without some estimate of 
the different times and circumstances in 
which the two poems were composed, Heyne 
has the merit of having pointed this out in 
1753, but we have no History of Alexandrian 
Literature to refer to until we come to that 
of M. Conat in 1882. Meanwhile several 
French writers in a vagve and rhetorical 
way had made the usual comparisons, and 
even Sainte-Beuve by detaching the episode 
of Medea in Bk. iii. had injured the repu- 
tation of the rest of the Argonautica. It 
was considered enough to say generally that 
Apollonius had inspired Vergil in the fourth 
book of the Aeneid. In 1838 however 
Patin, in the introduction to a course of 


176 


lectures on the Latin poets, had characterized 
the relationship between tho two with much 
more accuracy, and what M. de Mir- 
mont gives us is, in a great measure, an 
expansion of the view there shortly stated, 
though one rather less favourable to Apollo- 
nius. Neither M. Girard, who has written 
on Alexandrianism in his Etudes sur la Poésie 
Grecque (1884), nor M. Conat, in the work 
above mentioned, has gone into this question, 
as it did not enter into their plans, and the 
same remark applies to Susemihl’s elaborate 
and valuable //istory of Greek Literature 
in the Alexandrian Time (1891). 

M. de Mirmont devotes some pages to the 
aim of Vergil in writing the Aeneid, about 
which there is now little dispute. Vergil’s 
aim was the glorification of Rome and the 
exhibition in the person of Aeneas of the 
ideal Roman character. Vergil had the 
great advantage of being able to take a sub- 
ject that was at once national and religious. 
‘T’auteur de |’Encide,’ says our author, ‘a 
eu l’heureuse fortune de composer son 
épopée & un moment ot il pouvait combiner 
Vhistoire et la légende, et le talent de faire 
de cette combinaison un ensemble har- 
monieux et parfait.’ The definition given 
of the Aeneid is ‘une patrie fondée par un 
héros pieux.’ Now both these sources of 
inspiration were wanting to Apollonius. 
His epic could not be patriotic, because 
there was no national sentiment among the 
Greeks in Egypt: it could not be religious 
because there was then no serious belief in 
gods. 

T think however M. de Mirmont goes too 
far when he asserts that an epic poem must 
be national and religious or it cannot de- 
serve the name. No doubt the greatest 
epics of the world have been inspired by 
one or other of these exalted motives, but 
to require both would, I fear, exclude all 
but Homer and Vergil from the list of the 
epic poets. It is however, according to 
Aristotle, sufficient to constitute an epic 
poem that the theme be dignified, that it 
be in the way of narration, and that it 
have a certain unity. Now it cannot be 
denied that the Argonautica of Apollo- 
nius satisfies these conditions. The unity, 
it is true, ismerely mechanical and not vital. 
The voyage and adventures of the heroes 
form the loose bond of union between various 
episodes. It would be much easier to split 
up a good deal of the Argonautica into 
separate lays than to split up Homer, and 
this would probably have been done long 
ago by German scholars if we did not happen 
to know too much about the genesis of the 


THE CLASSICAL REVIEW. 


Argonautica, For this reason M. de Mir- 
mont is of opinion that Apollonius was 
really following the ‘precepts and practice 
of Theocritus, Callimachus and the Alexan- 
drian school generally, as expressed in the 
saying péya BiBdiov péya Kaxov. I cannot 
share this opinion, nor do I think that 
Apollonius would thus have defended him- 
self. The famous quarrel between Calli- 
machus and Apollonius would hardly have 
become so embittered if ‘it had been capable 
of adjustment by a little explanation. It 
is generally thought, and in my opinion 
rightly thought, that the Argonautica was 
partly a protest against the powerful influ- 
ence of Callimachus, who aimed at replacing 
long narrative poems ‘par des récits de 
courte haleine.’ 

This is not the place to discuss the real 
meaning underlying the Argonautic legend 
—if there is anything to find out—but I 
may say that it is here maintained that the 
original object of the voyage was to expiate 
the wrong done to Phrixus either by bringing 
back his body or giving rest to his soul by 
the performance of the proper rites, and 
this certainly is alluded to in Pindar’s Fourth 
Pythian ode, and obscurely once or twice in 
Apollonius. Now, as Hesiod . knows 
nothing of this, but speaks of the expedi- 
tion as an adventure of Jason for the pur- 
pose of carrying off Medea after the per- 
formance of the prescribed labours, the 
conclusion is drawn that by the time of 
Hesiod the legend of Phrixus had become 
separated from that of Jason. But surely, 
from the analogy of most other legends, 
the natural inference is the other way—that 
in course of time the legends of Phrixus 
and Jason, originally separate, became fused 
together. Inthe Argonautic legend Jessen 
(Prolegomena in .catalogum Argonautarum. 
Diss. Berol. 1889) finds traces of legends 
belonging to the Minyans, Boeotians and 
Argives respectively. Similarly the heroes 
seem to have originally been Minyans—and 
that remained their collective name—but as 
the tale grew more famous heroes were 
added from various parts of Greece. 

M. de Mirmont finds that the Argonautica 
is an encyclopaedic work—a mélange of the 
mythological, historical, geographical, and di- 
dactic poem, and that, though epic in form, it 
also, as Patin says, combines elements from 
lyricand dramatic poetry. From its nature, 
it could not be and was not intended to be 
popular, and was written only for the learned 
few of the Museum. The Roman writers 
shortly before the time of Vergil, viz. 
Catullus (in his Zpithalamiwm), Helvius 





THE CLASSICAL REVIEW, 


Cinna, Licinius Calvus, and others, had 
done little more than imitate the Alex- 
andrians, but Vergil saw that this endan- 
gered the originality of Roman poetry and 
renounced it. As his genius was far greater 
than that of Apollonius, he can be com- 
pared and contrasted only with Homer in 
all the higher qualities of his art, and he 
only makes use of Apollonius to enrich his 
poem by certain elegances and refinements 
of thought and expression. But, as Vergil 
uses his material like a master, not a servant, 
he shows his superiority by improving upon 
his model and never sinks into the mere 
copyist. Between Homer whom he imitates 
and Vergil by whom he is imitated, A pollonins 


177 


resembles an earthenware pot between two 
brazen vessels. This seems to me to be sub 
stantially the conclusion which M. de 
Mirmont reaches, and it can hardly be 
doubted that it is on the whole the correct 
one; and in spite of minor differences of 
opinion, it must be recognized that he has 
done good service by this interesting disser 
tation and traversed ground not previously 
occupied. In a word, in reading Apollonius 
and Vergil side by side we must bear in 
mind not only their similarities but also 
their much greater dissimilarities, so that 
they can be compared only with important 
reservations. 


R. C. Seaton, 


CRUSIUS ON THE DELPHIC HYMNS. 


Die Delpiischen Mymnen. Untersuchungen 
tiber Texte und Melodien von O. Crusivs. 
Gottingen: Dieterich’sche Verlags-Buch- 
handlung. 1894. S8vo. Pp. ii, 168. 
As. 6d. 


THis treatise is devoted to the group of 
hymns that were published by MM. Henri 
Weil and Théodore Reinach in the Bulletin 
de Correspondance Hellénique, vol. 17, pp. 561 
—610, and plates 21, 22, A number of 
questions have been raised by all of . them ; 
but I shall not go beyond the principal 
piece—the so-called Hymn to Apollo—and 
the principal point about that piece, namely, 
the manner of transcribing the musical notes.! 
Dr. Crusius admits that there is nothing 
in the notes themselves to show whether 
they refer to a chromatic or an enharmonic 
scale. He assumes that they must here 
refer to a chromatic scale, as the enharmonic 
had gone out of use before the date at which 
the piece was written. And then he jumps 
.to the conclusion (p. 104) that they refer to 
the chromatic scale with intervals of half a 
tone apiece. M. Reinach, however, took care 
to mention that we have the choice of three 
chromatic scales at least, namely, the chroma 
toniaion with its intervals of half a tone 
apiece, the chroma malakon with intervals of 
two-thirds of a tone, and the chroma hemiolion 
with intervals of three-eighths. He decided 
in favour of the toniaion, partly because it 
was the only one that could be rendered on 
1M. Reinach’s transcript has been reprinted by 
Mr. Monro in his Modes of Ancient Greek Music, pp. 
134 ff., without notice of the difficulties. 
NO. LXXVII, VOL, IX. 





the piano, and partly because it was the 
easiest and thus most suited to a hymn. 

Whether the scale was enharmonic or 
chromatic, it certainly was Phrygian. And 
this is admitted by M.Reinach and by Dr. 
Crusius also; so that both these high 
authorities are committed to the view that 
the chroma toniaion could be associated with 
the Phrygian chromatic scale. 

In the chroma toniaion, with its intervals 
of half a tone apiece, the note named 
paranété synémmenén stood two half-tones 
above the note named mesé, and therefore 
coincided with the note named paramesé, 
which stood a tone above the mesé. But in 
the Phrygian chromatic scale these notes are 
not the same, the paranété synémmendn being 
x, While the paramesé isu. And the only 
scales in which they are the same,are the 
®olian chromatic, lower olian chromatie, 
upper AMolian chromatic, and the so-called 
lower Ionian chromatic, which is simply the 
upper Xolian, an octave below its proper 
place. This seems to show conclusively that 
the chroma toniaion was confined to scales of 
the Molian group, and could not be 
associated with a Phrygian scale. 

Tn order to maintain his view, M. Reinach 
was obliged to treat « and cas a single note ; 
and Dr. Crusius has done the same—without 
a word of explanation. M. Reinach's 
explanation was that he regarded « and «as 
a pair of homotones. ; 

Now, the ancient notation is designed for 
one-and-twenty notes within the octave ; 
each of the seven original notes being 


followed by two supplementary notes. The 
N 


178 


notation for the instruments shows this very 
plainly, the signs recurring there in groups 
ofthree, F 1s A Ow D Ko Seca and 
so forth. In the notation for the voices, 
which is adopted in the hymn, « and 2 are 
the supplementary notes to mw, while vis an 
original note with » and 6 for its supple- 
ments. 

When the notes within the octave were 
reduced from twenty-one to twelve, nine of 
the supplementary notes had to be omitted. 
The result was that, while » had previously 
been followed by two supplementary notes, 
it now was followed by only one ; and this 
was known indifferently asx or A. Hence 
x and A became what Gaudentius! calls éudrova 
and Aristeides? calls cuudwvia.. But there 
does not seem to be the least foundation for 
the notion that «x and x could become a pair 
of homotones. 

Even if « and. were capable of forming 
such a pair, they could hardly form it here. 
As already stated, the paranété synémmenén 
and the paramesé are indicated by a single 
letter in four of the chromatic scales, and 
by two different letters in the rest. And 
this distinction would be meaningless, if two 
different letters could be used when the two 
notes were the same. 

One of M. Reinach’s reasons for adopting 
the chroma toniaion was that ‘ il est le seul 
qui puisse s’exécuter sur un de nos instru- 


1 Gaudentius, p..23, ed. Meibom. 
2 Avisteides, p. 26, ed. Meibom ; p. 16, ed. Jahn. 


THE CLASSICAL REVIEW. 


ments & tempérament’ ; and his transcript 
has the modern notation which is associated 
with the tempered scale. Dr. Crusius uses 
this notation also—pp. 152 ff.—and thus 
seems to be assenting to the notion that the 
tempered scale was known in ancient times. 

This notion cannot be based on any 
authority but Aristoxenos ; and there is a 
passage in Aristoxenos which disproves it 
altogether. He begins by saying that a 
fourth and a fifth together made an octave. 
Then he says that a tone was the difference 
between a fourth and a fifth; thus making his 
tone a major tone, not a mean tone. And 
then he says that a fourth contained two 
tones and a half. The tones being major 
tones, the so-called half-tone must have been 
(approximately) half a minor tone, So he 
followed the Pythagoreans in their division 
of the octave. 

That gives five major tones and two semi- 
tones of (approximately) half a minor tone 
apiece, in place of the five mean tones and 
two mean semitones, which together form an 
octave in the tempered scale. And if the 
chroma toniaion, had been used, its half-tones 
would have been the halves of major tones. 

No doubt, the discrepancy is very slight 
indeed. But it cannot safely be ignored. 
Greek music with the tempered scale would 
be as bad as Greek architecture with straight 
lines substituted for its subtle curves. 

Crcin Torr. 
3 Aristoxenos, pp. 45, 46, ed. Meibom. 





ON THE MANUSCRIPTS OF PROPERTIUS. 


Mr Hovsman in his polemic against my 
pamphlet On Certain Manuscripts of Pro- 
pertius has two objects in view, of which the 
connexion will not escape the observant,—to 
‘adjust,’ so he calls it, my ‘ partial estimate’ 
of the Holkham MS (L) of which the read- 
ings were there first published and discussed, 
and to demolish objections to theories of his 
own which were there brought forward. Mr 
Housman’s style of controversy is now 
familiar to the public, nor needs it char- 
acterizing by me ; and the present specimen 
seems fully to sustain its author’s reputation. 

He begins with the nine or ten cases 
where L has a better reading than the 
other MSS8,—N(eapolitanus), F (Lauren- 
tianus), D(aventriensis), V(aticanus); A 
(Vossianus) being deficient in this part of 
Propertius. TI had said (pp. 62 sq.) ‘a doubt 


greater or less according to circumstances 
must restuponall unsupported lectionsin any! 
of the manuscripts AFLDV.’ Little room 
for adjustment here, one would think. But 
Mr Housman goes through the list which I 
had collected in fairness to L and for the con- 
venience of my reader, shows that all in it 
may be due to accident or conjecture, and 
adjusts as follows ‘ L therefore brings! nothing 
new and true to the constitution of the text’ 
and again ‘ L has nothing good of its own.’2 A 
promising beginning this. 

1 Italics throughout are mine unless I specify 
otherwise. 


* Mr Housman says I ‘essay to base a conjecture 
on an unsupported reading of L at III iv 22.” What 
I essay to base it on is that reading taken together 
with the serious discrepancy media and sacra 
between the other MSS. He again perverts my 


argument on this passage in discussing II xxiv 46. — 


4 


THE CLASSICAL REVIEW 


Next comes a review of the evidence that 
L is an independent witness to the tradition 
of the family ® (the family of A, F, and,in 
part, of N), which is not all satisfactory to 
Mr Housman. Mr Housman, having accepted 
my conclusion, is welcome to pick and choose 
among my arguments; but against a con- 
troversialist who sets up dummy hypo- 
theses I must observe that by the end of 
this reply my readers will be able to judge 
whether I was not justified in neglecting 
theories which even Mr Housman does 
not venture to maintain. 

The next paragraph deals with the use of 
L in determining the tradition of 6. Mr 
Housman writes, p. 20: 


‘Where it (L) agrees with F, there the common 
reading of the two will generally be the reading of 
@; not always, for at II xxxii 8 they both have 
timeo, but generally. But where L dissents from 
F, that fact in itself tells us nothing about @ and 
shakes F’s testimony not a whit. . . The adherence 
of L to N is absolutely weightless as witness to the 
tradition of @: both steal from other sources, and 
one of the sources whence L steals may very well be 
N itself: F remains the sole untainted channel of 
the family tradition when A is absent.’ 

Of these three independent witnesses to 
the tradition of ® (F, L and N) F has 
obscured it by intrinsic corruption only ; 
but in L often and in N oftener it has 
been replaced by readings derived from 
outside. It is our duty then to ascertain 
from examining undoubted instances in 
what proportions L and N have the ® 
tradition or an alien one and apply the 
ratios to the doubtful ones. Suppose the 
ratio for L to be 80 p.c. and for N 70 p.c., 
then in at least 50 and possibly in 70 out 
of every hundred corrupt places in F’, the 
® tradition or traces of it will be preserved 
both in Land N. If it be urged that this 
is of no use to usas we cannot tell which these 
places will be, the answer is; here is a 
critic’s opportunity. He will closely 
scrutinize the MS indications and will for 
example conclude that at IIL xx 17 F's 
pignita means that pignera (LN) was in ® 
rather than pignora (DV) because @ for eis a 
far commoner error than 7 for 0.1 And he 
will prefer to hold that in these and similar 
eases (my p. 34) L has preserved with N the 
tradition which F has corrupted rather than 
hold that it has replaced some absolutely 
unknown reading by another possibly de- 
rived from N. But Mr Housman, though he 
To make unlikes like by confusing the circumstances 
seems to be out of many Mr Housman’s favourite 
paralogism. 

1 Mr. Housman uses the indications of F at II 
xxii 22 and elsewhere ina precisely similar way J.P. 


_ xxi 140, 


179 


allows the ‘adulterated’ and ‘untrust 
worthy’ N to count against both F and J, 
together (as at II xxxii 8), and again allows 
L to count with F against N, yet says that L's 
adherences to N are ‘absolutely weightless’ 
against F, Had the cabman of the story 
been a Propertian scholar, he would surely 
have called Mr Housman also a ‘ harbitrary 
cove.’ 2 : 

Mr Housman next says :— 


‘The adherence of L to DV is equally unimportant : 
the tradition of that family is seldom doubtful, so 
that a third witness is superfluous, especially so poor 
a witness as L.’ ; 


If he is criticizing me here, he has mis- 
conceived my reasoning. Whenever I speak 
of L conferring a corroboration upon 
readings of A (7.e. DV or their family), I refer 
to the doubt which overhangs the date and 
composition of their parent MS, as p. 63 
(n.) : ‘ We now know from the evidence of L 
that some readings of A are older than 
1421; and this is all we know for certain.’ 

Mr Housman finally directs us how often 
to cite L in a critical apparatus—‘ never...,’ 
he says, ‘unless it agrees with F and 
dissents from N, or else presents a reading 


2 In his note on peraeque III xxviii 9 (p. 27 n.) Mr 
Housman, angry at being told he has not attended 
enough to manuscript abbreviations, surpasses him- 
self. L (L writes e¢ for ae always) having peque, D 
(and seemingly V formerly) pareque, F paremgue and 
N per acquae and I having said the original of all 
was ‘neither ‘‘pereque” nor ‘‘pareque” but 
“neque ”,’ Mr Housman puts forth as ‘probable’ 
the peculiar hypothesis ‘that the readings of DVF 
are due to weque, which their common parent O had 
misinterpreted as pareque. Why! To deprive L 
of the honour of having preserved the common 
tradition more faithfully than DV and F. As to 
N’s per aequae, I had derived that from peque because, 
as the variation of ae and e¢ is unimportant and is 
disregarded here by Mr Housman himself, 
J.P. xxi 131 ‘per aeque, that is peraeque,’ there 
is no solid ground for separating N from the 
other MSS with which it agrees far oftener than 
it differs. But Mr Housman says ‘N derives scores 
of readings from an older source than O, and there is 
pot a hint that this source had peque rather than 
peracque.’ So, having provided one source older 
than O with gegue against L, he provides another 
with peraeque against me, Place aux ombres ! After 
contradicting my statement that N’s reading was 
no better than DV’s or F’s(F I added, speaking to one 
point at a time) by saying that ‘truth is more 
obscured by wrongly expanding the abbreviation “— 
let him add, to himself—‘ than by misdividing the 
word’? as if that were the test to apply toa seribe, he 
ends with this outburst: ‘On p. 23an equally base- 
less charge of neglecting abbreviations 1s brought 
r Leo and supported only by flat contra- 
diction.’ The wondering reader may be told that 
Mr Housman has discussed the place in question 
(II xxxii 8) in J.P. xxi 174 without referring to = 
abbreviation (¢ for ¢ibi), a compendium also i — 
ib, p. 129, on IV iii 61 and i. p. 141, on LV v 85: 


compare my pamphlet p. 23 and n. 1. - 


against M 


180 


peculiar to itself....If i were editing Pro- 
pertius I should mention L in about thirty 
places.’ On referring to the very much 
abbreviated apparatus at the foot of ‘my 
text in the Corpus I find there about sixty- 
five instances where ‘L agrees with F and 
dissents from N’ and about thirty-two 
where it has a reading, correct or interesting, 
‘peculiar to itself.’ Total about ninety- 
seven. Either then Mr Housman’s appa- 


ratus criticus would be of the most meagre ~ 


description or he has ‘adjusted’ L’s 
admitted claims to consideration by dividing 
by three. 

We now have introduced the real gist and 
matter of the controversy—the validity of 
Mr Housman’s own views upon other manu- 
scripts of Propertius. These are set out in 
the Journal of Philology in what I had 
called (p. 61) a ‘ brilliant triad of articles,’ 
from which I have myself learnt much and 
which after all deductions will remain a 
solid and durable contribution to Propertian 
studies. One might have thought that this 
constituted a necessity for criticizing the 
errors which it contains. Not so thinks 
Mr Housman. This ‘necessity’ was ‘ purely 
subjective.’ Neither he nor his work must 
be profaned. He ‘had hoped he had done 
with this matter for a long time to come.’ 
But I come ‘hacking at the fence’ of his 
paradise, I ‘let the boars in among the 
‘ crystal springs,’! and ‘like Nehemiah’s 
builders’ he must arm in defence of his 
Holy City. And then the injury to his 
feelings! ‘If it were not for the humour 
of the situation, I might well resent the 
tone of placid assurance in which J, who 
think before I write and blot before I print, 
am continually admonished by the author of 
this pamphlet.’ Precisely. yéAwra 59 TON 
EME év tots Adyous drédarge Plato Theaetetus 
166 A. There 7s humour in the situation ; 
but not what Mr Housman supposes. The 
humour is this, that Mr Housman, who has 
rated half the scholars of Europe, should 
himself be so sore at reproof and should 
expect his indignation to move the readers 
of the Classical Review. 

The first issue is the trustworthiness of 
F. Mr Housman gives out that he upholds 
it against myself who said ‘we cannot trust 
it unconfirmed.’ Let us collect his own 
- statements. ‘F, whose scribe was a most 
ignorant man, is defaced by a hundred 


1 ‘For no discoverable reason unless it is the hope 
of boasting ‘“‘liquidis immisi fontibus apros”’ (p. 22). 

2 Of the contumelious reference here to my dis- 
cussion of III xiv 19 (p. 184 below) I intend to make 
Mr Housman ashamed. 


THE CLASSICAL REVIEW. 


blunders which were not in the exemplar 
whence it was copied’ J.P. xxi p. 181. ‘F 
is carelessly written by an illiterate man’ 
ib. xxii p. 128. ‘I attach little weight to 
F’s unsupported readings from I 7 1 to 
IT 4 63’ Cl. Rp. 23 as. S When ‘at TY 
vii 25 DVNL give fissa and F fixa, no 
conclusion can be drawn-respecting ®. May- 
be ® had fissa and F, as often, blundered ; 
maybe ® had five and L, as often, stole from 
the other sources; no one can say,’ ib. p. 21 
(in other words between IT «63 and the end, 
where F stands by itself, we cannot trust it 
unconfirmed). Thus does Mr Housman 
maintain the value of F. What then is 
there left for me to do? What need for 
homicide when the suicide is so complete? 
He even subverts the good faith of F where 
I had left it intact. I will show him how. 
At IIL iii 11 F has correctly /ares and Mr 
Housman has conceded that /acres was in ®. 
Now (on p. 25) Mr Housman says, ‘ When 
critics’ [that is Mr Housman] ‘find a “uox 
nihili”’ [that is acres] ‘ altered into a Latin 
word’ [that is /ares] ‘they do not call it 
confounding two letters but they call it a 
conjecture.’ 

To turn to Mr Housman’s replies to my 
criticisms, I had said that in many of Mr 
Housman’s instances, where the difference 
between the variants was very slight, it 
seemed more likely that F was ‘accident- 
ally right, has in fact blundered into a 
correction than that one and the same 
corruption should have appeared indepen- 
dently in all the other manuscripts.’ After 
stating this view (about which he says 
himself that he is ‘quite willing to admit 
that this might sometimes happen’) I 
considered some of the examples to which 
he attached exceptional importance: ‘ Among 
our examples there are very striking tokens 
of integrity ; the recondite /unarat, Tiburtina 
retained despite the metre, the form Tarpetia 
disguised but thinly, the form cwmba 
preserved’ J.P. xxi 197. And, imagining 
that it was because /unarat was ‘ recondite ’ 
and because Ziburtina violated the metre, 
that Mr Housman thought their retention a 
very striking token of integrity, I argued 
that this was not striking in the case of F, 
which abounds in strange words and unme- 
tricallines. Mr Housman’s ‘head goes round’ 
at being thus understood ; and he presents 
me with the following explanation: ‘A 
striking case’ {in which F has abstained, 
from the absence of motive, from altering | 
strange words and unmetrical lines] ‘is a 
striking token’ of F's ‘integrity’ [or of F’s 
abstaining, from the absence of motive, from 


— 
a 


_ family are found in F alone’ (p. 22). 


THE CLASSICAL REVIEW, 


altering strange words and unmetrical 
lines].' So exit Mr Housman by the same 
hole he went in at. I can only say that, if 
lunarat is not a ‘striking’ case for the 
reason supposed, it is striking for nothing 
else ; for I take it from Mr Housman J.P. 
xxii p. 98 that ‘/unarat F’ and ‘limarat 
DVN’...‘are palaeographically almost iden- 
tical’; and so that here too F may be 
accidentally right. Ziburtina on the other 
hand I believe to be actually wrong.? 

Mr Housman lays stress on the number, 
now reckoned at 21, four of which however 
are also in L,? of F's good readings, and 
leaves the reader to infer that they are 
certain. The re-collation of about one-tenth 
of the portion from which his good readings 
were drawn has shown that two (now 
omitted) IT xi 1 ne and xxxiii 43 absentis were 
mistakes of Baehrens. What a woeful re- 
duction in the total there would be should 
anything like this ratio be maintained ! 

Mr Housman next argues that to shake 
the solitary witness borne by F will ‘land’ 
me ‘in a conclusion ’ at which I do not desire 
to arrive. ‘To depreciate F is to depreciate 
the whole family © (as distinct from N) ; for 
all the important readings peculiar to that 
Was 
a question ever begged more openly? All 
that we know of ® is gathered from the 
agreements in AFL (and N). When F 
stands alone then, to quote Mr Housman 
again, ‘no conclusion can be drawn respecting 


1 The matter I add in square brackets is taken 


almost verbatim from Mr Housman, p. 21. 


2 In this connexion I had written that ‘Mr 


_ Housman’s advocacy of the value of F’s isolated 


witness involves him in a curious inconsistency. He 


_ follows Baehrens in maintaining that ‘‘ A is the most 


faithful representative of its family ”’...It certainly 
then ‘‘ happens curiously ” that in the poems in which 
we have both A and F, A should give of itself but 
one true reading...and F three or four,’ which Mr 
Housman thinks after all may be an accident.’ His 


defence, with Dick, Tom and Harry to the rescue, is 


“wherever A gives a true reading that reading is also 
given by F or by N’ (hisitalies) ‘or by both.’ Those 
acquainted with the laws of probability will still 
think it curious that the faithful A, unlike the less 
faithful F, should never be right by itsc/f. But the 
‘imputation of ‘inconsistency I will withdraw, as I 
can find no evidence that Mr Housman has any such 
acquaintance. 
__ * Mr Housman’s reference to these instances gives 
curious indication of the bias and confusion in his 
mind. ‘In four of them F is now confirmed by L, so 
Dr Postgate must confess that in those four I was 
right,’ p. 21. Why, when did I say otherwise? I 
‘Rever criticized his views on F till the pamphlet he 
has attacked ; from it alone docs he derive these 
£onfirmations and even. the remark is ‘plagiarised’ 
from me, p. 29. He should have written ‘In four 
of them F is now confirmed by L, as I have learnt 
from Dr Postgate.’ 


181 


ewe ey ; 
®.’ Readings then peculiar to A, F, or L 
must be put aside in considering the family 
®: being uncertain they cannot add to our 
knowledge nor must affect 
judgment.! 

Another issue is the authority of DV 
(or A). Mr Housman writes, p. 22: 


they our 


‘It was easy to foresee that the next writer on 

Propertius’ MSS would disparage DV. Baehrens had 
(disparaged N, Mr Leo had disparaged O, Mr Solbisky 
had disparaged AF, I had defended one and all! : 80 
to disparage DV was the only way left of being 
original.’ 
Mr Housman has nowhere insulted the in- 
telligence of his reader more openly than 
here. Why, DV and AF are O, Mr Leo has 
disparaged both; and when Mr Housman 
thinks another scholar has taken half of a 
published theory of his own, he does not 
call that ‘originality,’ he calls it plagiarism. 
But what is to be said when it is observed 
that at the head of my. discussion I have 
acknowledged my obligations to Mr Leo in 
these express terms !—-‘ The integrity of DV 
has been already impugned by Leo kh. Mus. 
35, 442, and on good grounds as the evidence 
to be adduced will show.’ Mr Housman 
however must provide me with a motive ; and 
a craving after originality would do. 

My distrust of DV arose differently. In 
the course of that close and constant scrutiny 
of the manuscript variants, which the pre- 
paration of a text entails, doubts of their 
honesty formed in my mind. This impression, 
which I received to nothing like the same 
extent in the case of the other MSS,° grew 
continually, and made it necessary to 
examine the evidence, which is collected on 
pp. 69 sqg. of my pamphlet. As dishonesty 
has various degrees from deliberate falsifica- 
tion to culpable carelessness in copying, and 
as with the number of proved interpolations 
grows the probability that the doubtful 
cases are interpolations too, I have divided 
it into cases of interpolation and of grave 
suspicion. Mr Housman now charges me with 
using ‘divers weights and divers measures 
and describes my procedure as * the good old 
rule, the simple plan, of “ heads I win, tails 
you lose.”’ I ‘denounce ’ in one manuscript 


4 The only ‘important’ unsupported readings are 
those readings which, being ap yarently a ser 
diverge too much from the rest to be due pn _ 
and those readings which reveal what the scribe va 
before him. ‘To this latter class, of be te 
spoken (on p. 35), belongs F's Tarpetia whic re 
Mr Housman says, indicates TJarpetie in ie 
exemplar. 

5 This is why I have 
the similar evidence ’ against #. 
has refrained from collecting it. 


‘refrained from collecting 
Mr Housinan also 


182 


‘the same offences’ that I ‘extenuate’ in 
another. Mr Housman’s own method is 
simpler still, it is to examine a fraction of 
the evidence, selected by himself, and to 
ignore the rest; a ‘transparent iniquity ’ 
which the reader will not fail to detect. 

Though the test is such an unfair one, 
I will not decline it, and refer first to 
II xxiv 45 sq. ‘iam tibi lasonia uecta est 
Medea carina | et modo ad infido sola relicta 
uiro, so DV, seruato N, om. FL. I wrote 
that ‘one of the two (readings) must be an 
interpolation. In deciding which, we ask 
first which gives the easier construction ; 
and the answer is ab infido; and secondly 
which presents the more obvious sense ; 
and the answer, as we see from another 
supplement “fallaci”’ in D, is again ab infido.’ 
Mr Housman retorts with the dilemma 
(p. 26) that ‘if the supplement ‘fallaci” 
shows that ab ifido presents the more 
obvious sense, then it equally shows that 
seruato gives the easier construction. If it 
does not show that serwato gives the easier 
construction, then neither does it show that 
ab infido presents the more obvious sense.’ 
Pretty, but for the flaw in ‘equally’! 
‘fallaci’ does not equally show that serwato 
gives the easier construction because the 
greater obviousness of the sense may have 
overpowered, as I believe it has, the greater 
ease of. the construction. After this ebul- 
lition of fallacy Mr Housman addresses 
himself to the point as follows: ‘Nor does 
there appear to be any tangible difference 
in obviousness between the two senses or in 
ease between the two constructions.’ No 
tangible difference, that is, between the 
constructions of the ablative with ad and 
the dative with a passive participle ; none 
in obviousness of sense between that of 
saying the lover who has left Medea lone 
was ‘faithless’ and that of saying that she 
‘had been his saviour.’ Let the reader 
judge whether I have not rightly argued 
that A’s reading has two marks of an inter- 
polation. But Mr Housman, having once 
swallowed the explanation of Baehrens 
that the scribe’s eye slipped from modo to 
infido, refuses to disgorge it, and though 
forced to allow that such slipping is not the 
sole cause of omission, insists that the 
‘rational’ inquirer must recognize it here. 
The answer is obvious. Baehrens’ explana- 
tion is palaeographically the easier one ; 
but other considerations are against it. 
Nor am I inconsistent in adopting the 
explanation at III iv 22 (above p.178 n.) 
where those other considerations are in its 
favour. 


THE CLASSICAL REVIEW, 


Of IL.xxxiv/83~ andi HT vit LINE sam 
briefly that /audabat for ludebat is suspicious 
because its sense is easier, and externas 
for hesternas because L indicates the 
intermediate stage hexternas; and of II 
xix 26 that if Mr. Housman can persuade 
any one that pedes for boues comes from 
I xx 8 or IV xi 16, he. is welcome to his 
proselyte.1 

Mr Housman accuses me of Iago-like 


-malignity in imputing fraud to DV. So 


the reader may be surprised to hear that 
I have deliberately omitted from my list 
a passage where Mr Housman, on insuflicient 
grounds, has pronounced them interpolated : 
IT xiii 47 (JP. xxi 140) ‘ me-minisset 
[DV] is a metrical correction of minisset.’ 
But amazed he will not be, as soon as he 
observes that in the very case (I xix 10) 
selected by Mr Housman as a ‘specimen’ of 
my ‘ uncharitable imputations’ he has him- 
self declared their reading an interpolation 
and rebuked the scholars who have built 
upon it: ‘ werberat is soon explained as an 
attempt to correct the easy mistake ueberat : 
there is no cause or defence then for 
Baehrens’ suspicion peruolat, or for Mr 
Rossberg’s surprising werterat’ (ib. p. 183). 

A word on the age of A. Mr Housman 
had written in J.P. xxi 180 n. ‘it was 
probably earlier than 1400, certainly not 
much later.’ These words, especially those 
in italics, mean that he was then of opinion 
that the approximate date of A was about 
1400: they are inconsistent with a date 
one or two centuries earlier. Now he says that 
that was ‘merely his modesty; he claimed 
no more than he wanted for the point he 
was then discussing.’ So when an unex- 
pected and unpalatable use has been made 
of this statement, his ‘ modesty’ throws it 
to the winds, and he declares that ‘ nothing 
ties down A to the 14th century or near it: 
it can easily be older than any date yet 
assigned to N.’ I tell him then that of 
the date of A he is wholly ignorant; it can 
easily be later even than L.2 All he knows 


1 Mr Housman errs in supposing I think D inter- 
polated at II xxx 36 from I i 143; but I see now 
I did not make my meaning clear. The phrase 
(p. 87) ‘the scribe allowed himself to write’ (not 
‘wrote’ or ‘interpolated’) I intended to be taken 
with the preceding, not the following context ; and 
to exemplify the way in which the scribe allowed his 
mind to work upon his subject. 

° There is nothing against this but Baehrens’ 
opinion that D’s date is 1410-20. As Baehrens 
appears to have mis-dated all the Propertian manu- 
scripts except I’, whose first owners’ names are written 
inside, this counts for nothing. As to V, which he 
places at the end of the 14th century, MM. Steven- 
son fils, Maurice Faucon, and Pierre de Nolhac agree 






} 


a 





THE CLASSICAL REVIEW, 185 


is that some of its readings are earlier than 
1421; and this he has learnt from me. 

Lastly Mr. Housman disputes my esti- 
mate of the Neapolitanus. Before replying 
I concede to him (p. 28) that the argument 
I based on III v 35 plaustra Bootes to show 
N was an independent witness to the read- 
ings of O, the supposed common parent of 
A and@, isfaulty. I ought to have treated 
this O throughout for all that at present it 
is, 2 convenient algebraical symbol for the 
common readings of AFLDV. 

Summing up (p. 73) I had said first that 
it was ‘best as being the oldest of our 
witnesses.’ Mr Housman comments (p. 23) : 


‘But age is no merit. . . Till we have examined 
two rival MSS, we presume that the older is the 
better. When we have examined them, we judge 
them by their contents. . . Useless then to call the 
Neapolitanus ‘“‘best as being the oldest of our 
witnesses,” unless you can keep it out of our reach.’ 


This appears to mean that a manuscript 
is to be appraised by internal evidence only ; 
or that codices of the first, fifth and fifteenth 
centuries, if they each have equally good 
texts, are of equal authority. Unless Mr 
Housman maintains this absurdity, he must 
imagine that I believe the oldest codex to 
be always the best, although I wrote (p. 65) 
‘Ceteris paribus, —Ceterts paribus, I repeat 
—‘the older testimony must be believed.’ 
The next point involves the composite 
character of N. I had said that ‘it was 
best again as the one that presents the 
greatest amount of truth with the smallest 
amount of falsehood.’ Mr Housman, un- 
able to advance a step without ascribing to 
his opponent: his own vices of reasoning, 
says ‘Then if I set a clerk to copy out the 
Teubner text the result will be in Dr 
Postgate’s opinion a still better MS than 
the Neapolitanus, because it will present a 
greater amount of truth with a smaller 
amount of falsehood,’ as though, forsooth, 
the composition of the Teubner text were 
not matter of knowledge and that of the 
Neapolitanus matter of conjecture! Mr 
Housman writes of conflated manuscripts 
with a strange appearance of passion. ‘N 
adulterates its text.’ N and L ‘steal from 
other sources.’ This insinuation of dis- 
honesty is quite beside the mark. Manu- 
scripts were the books of ages that preceded 
printing, and it was the producers’ business 
to gather up all the truth they could from 


that this date is at least half a century out. And 
Mr E. Maunde Thompson’s opinion, based on the 
photograph in Chatelain’s Paléographie Latine, 1s 
that V was probably written ‘after the middle of the 
15th century.’ 


So 


the sources at their disposal. And well for 
us that they did, or our texts would be 
much worse than theyare. The ‘integrity’ 
which Mr Housman values so extra 
vagantly Was an accidental merit, due 
to the fact that only one source was 
accessible to the scribe. When therefore, 
to use his own phrases of N, Mr Housman 
‘rejoices’ to have a MS of the integrity of 
F, he finds ‘a wrong vent for that joy’ in 
extolling F; ‘the proper vent is to thank 
providence’ that it did not vouchsafe to F 
the means of correcting its errors. The 
value of A and F’s integrity is just this, 
Through it we can restore with some cer- 
tainty the readings of their immediate 
source (call it P). But it tells us nothing 
about P or P’s relation to ®. For all we 
know P may have been a conflated MS and 
this the explanation of some or all of its 
differences from N. 

With the next paragraph which deals 
with beings who believe N to be the best 
MS because it has no near relation to 
control its witness, I have no concern. The 
view is not mine, though Mr Housman 
makes a quotation from me the text of the 
paragraph. And the speculation what 
scholars would or would not hold were the 
manuscripts of Propertius other than they 
are, is an airy field where Mr Housman 
may disport himself at will. 

The next question is the relation of DV 
(A) to N. Baehrens asserted that their 
common readings were derived by N from 
A. This was reasonable enough in Baehrens, 
who placed A in the 14th century and N 
in the 15th; for conflated 15th century 
MSS of Propertius abound. It is not so 
reasonable in Mr Housman, who knows no 
more about the date of N than he knows 
about the date of A.’ ‘This ignorance 
vitiates the only two arguments he advances 
for his hypothesis. The first is: ‘W hat 
perversity, then, in order to avoid assuming 
that the MS known to have blent two 
elements has blent a third as well, to assume 
that the MS not known to contain more 
than one element contains three !’ The 
second is this: ‘The simplest hypothesis 
which will account for any given facts 1s 
held to be the likeliest hypothesis ’—a state- 
ment illumined by a reference to the su- 
periority of Copernicus’ account of the 
planetary system to Ptolemy's. The like- 


1 The starveling theory then is his own, not that 
of Baehrens, on whom he would father it. 
2 Mr Housman vapours here as esa — 
‘ardinals rejected the simple accoun 
College of Cardinals rejected t in 
aa it seemed to threaten Holy Writ. Dr Post- 


184 


liest hypothesis is however that one which, 
whether simple or otherwise, best explains 
the facts. When Mr Housman has ascer- 
tained these, when he knows the date of A 
and the date of N, we will talk about the 
number of elements and on the simplicity of 
hypotheses. Mr Housman objects that a 
theory similar to what I have suggested for 
the resemblances of N and A might be set 
up against the account I have accepted for 
those between N and ®, and that then 
my lips would be sealed, as ‘there is not a 
pin to choose between them.’ Mr Housman 
mistakes. The immediate progenitor of A, F 
[and L] is referred by Baehrens to about 
the middle of the fourteenth century ; but 
I think it was probably earlier! The 
indications appear to suggest a 13th century 
MS. @ was earlier still, and may well have 
been in existence at the earliest date to 
which N can be assigned. Thus the 
inference (I should call it no more) that 
N has borrowed from ® stands on quite a 
different footing to the hypothesis that N 
has borrowed from A. But ‘observe that 
the only one of the two which occurs to 
Dr Postgate is the one which jumps with 
his own prepossessions: I had considered 
and rejected both before ever I set pen to 
paper’ (p. 24). The comparison is crushing 
if Mr Housman be infallible: otherwise it 
has merely an autobiographical interest. 

We now come to the interpolations in N 
which I had said were much fewer than 
those in A. To show that Mr Housman was 
unfair to N I took first the passage which 
Mr Housman chose himself for the first 
place among his examples of N’s dishonesty. 
At III ii 1 sq. ‘Orphea delenisse feras et 
concita dicunt | flumina Threicia sustenuisse 
(Ayrmann) lyra’ detenuisse FV, detinuisse 
N, te tenuisse D, appear to be all descended 
from detenuisse. Now here, according to Mr 
Housman, changing / to ¢ and inserting w, 
thus making the strange -/enisse into the 
familiar -tenwisse, is accident, pure accident ; 
‘aslight and honest error.’ But writing tin 
for ten is quite another thing; ‘a bad attempt 
at amending.’ It is nothing that ¢ and ¢ are 
constantly confused in our MSS; nothing 
gate rejects the simple account because it is dero- 
gatory to the scarce less sacred Neapolitanus’ ! 
Thus does the habit of intemperance in language 
destroy the perception of proportion and even the 
sense of humour in its victim. 

1] think his date for A (1360) also somewhat 
too late. [Since the above was printed I have 
received Mr E, Maunde Thompson’s opinion based 
on the page of A photographed in Chatelain’s 
Paléographie Latine. He is ‘decidedly of opinion 
that the Vossianus was written about 1300 and just 
as likely before it as after it.’] 


THE CLASSICAL REVIEW. 


that the scribe’s eye might have been caught 
by the sustinwisse in the line below ; nothing 
that this estimate of a change of spelling 
would sow ‘interpolation’ broadcast over 
our manuscripts ; nothing that at II xxx 26 
the same MS shows no dislike to this very 
form.2 It is ‘a “uox nihili” altered into a 
Latin word,’ and therefore, by the critic’s 
definition, a conjecture. But here observe 
a curiosity. Mr Housman must needs add 
that ‘the term wox nihili’ (mine, in giving 
the point of view of the copyist of D) ‘is 
somewhat too harsh,’ ze. that